With speech and language has come the development of some notable differences between the left and right sides of the brain. In most other animals, the two halves are mirror images of each other. They not only look the same, but the right brain does for the left body what the left brain does for the right body. Not so in man, for speech and language are usually only on one side-the left in most people.

Knowing on which side language is located is obviously crucial for the neurosurgeon, since this is an area he must certainly avoid damaging. In Neil's case, the side of his brain which housed language was determined during a test done several weeks before the operation. In that test, a short-acting barbiturate drug was injected into the artery that supplies one side of the brain, temporarily putting to sleep just that one side. This was done while the patient was busy naming the objects shown by a slide projector. After one side of the brain was briefly put to sleep, there was a rest period and then the other side of the brain was similarly put to sleep. The side anesthetized when the patient stopped naming objects was the side housing the language portions of the brain.

An easier but less reliable method of telling which side houses language is to use a "dichotic listening test": two words differing from one another by only one sound are simultaneously presented to the two ears. After several sets of these word pa' rs, the patient repeats back the ones he heard. The ones he remembers are usually those presented to the ear opposite to the side of the brain housing language; thus, if "peace" is presented to the right ear at the same that "cease" is presented to the left ear, the patient will tend to report "peace" if he is left-brain-dominant for language. This kind of test works for reading too (and is called the "dichhaptic test"). The left brain initially gets all of the visual input from the right half of the visual field, and vice versa. If words are briefly flashed on the screen to the left and right of a point on which the patient is fixatl ng, the word on the right side will be more easily recalled by a patient whose left brain is language-dominant.

These and similar tests indicate that all language output comes from left brain in most people. The left side is also heavily involved with the ability to read, manipulate grammar, and regulate the sequence of movements of the face and hands. The right brain, in most people, has the ability to manipulate things in space, follow maps, remember shapes and faces and musical tone sequences, relate clothes to body image-it even seems to house the body image itself Thus, the "average" human has a verbal left brain and a visualspatial right brain.

But these abilities are not "lateralized" in the same way in everyone. Left-brain damage disturbs language in about thirteen people for every one person with language disturbance after right-brain damage. The ability to manipulate clothing In relation to one's body, as in getting dressed, is altered by right-brain damage in five people for every one person with dressing problems after left-brain damage. Right-brain damage interferes with the ability to manipulate objects in space, or follow routes, only about twice as often as left-brain damage. So it is not simply a matter of language on one side and visual-spatial on the other side-the mix is highly variable. Some individuals have both language and some (or all) visualspatial functions crowded onto the same side. But we do not know the consequences of this for function, except in the most extreme cases.

The degree of lateralization depends in part upon one's sex. Males are more likely to have visual-spatial abilities strongly lateralized to the right brain than are females. Failure to develop this strong lateralization of visual-spatial abilities has been identified in some boys with developmental dyslexia, a condition in which an otherwise intelligent child has great difficulty in learning to read. Dyslexia has a genetic component and is most often seen in males. These dyslexic boys have language in the left brain, like most people, but have visual-spatial abilities on both sides, unlike most males. This raises the possibility that the left language areas are squeezed by the left visual-spatial areas, to the detriment of reading.

There is a rough correspondence between handedness and this "lateralization " of language. Most, but not quite all, right-handers have language in the left brain. Left-handers, who are 15 percent of the population, are another matter. More than half of them also have language on the left, bringing the total of left-brain language up to 93 percent of the population. About half of the remaining left-handers (3 percent of the population) are true mirror-image reversals, with language in the right brain.

The remaining left-handers have language on both sides of the brain. This particular mix seems to be a less than optimal arrangement, since these people seem prone to language disorders, particularly stuttering. There are cases in which a stroke, involving only one side of the brain, has cured a lifelong stuttering problem, suggesting that stuttering occurs when the two sides are both involved with language and have difficulty in coordinating their act.

The very young show less evidence of lateralized functions; indeed, they seem to have the ability to move functions around following brain injury. The earlier the damage occurs, however, the better the chances for this compensation to occur;. after about the age of five, such flexibility is lost. For example, there is a congenital disorder in which abnormal blood vessels develop on just one side of the brain. If this condition is left untreated, severe seizures occur which not only affect the abnormal side but spread to the normal side and interfere with its development. To save the good half of the brain, neurosurgeons remove the abnormal hemisphere; the operation usually is done before six months of age.

Spotting the anatomical differences between left and right halves of the brain requires a view of the top of the temporal lobe, achieved by cutting the brain below the line shown at top in the side view. Looking down upon the cut brain (bottom), the infoldings of the rear portion of the temporal lobe are seen to differ between left and right sides (arrows). The dotted area is called the "planum temporale" and is much bigger (in most people) in the left brain. The function of this area is not well established, as stimulation mapping is usually prevented by its buried location, but the large left planum temporale is adjacent to the language cortex. The auditory-receiving area (see figure on page 9) is just in front of the planum temporale.

If it was the right hemisphere which was removed, language develops normally. In those who lost left hemispheres, speech develops, but not fully. They are considered quiet children, who use language only when necessary and then with a reduced range of grammatical expression. They tend to talk in the present tense; more elaborate constructions are beyond them. For example, when asked to repeat back the sentence "Wasn't the poor cousin helped by the old lady," one nine-year-old who had had the left hemisphere removed shortly after birth said, "Wasn't by the cousin helped by the old lady," while another recalled this sentence as "Wasn't the poor cousin . . . helped the old lady." Children with right hemisphere removals make few such grammatical errors. Apparently there is something wired into only the left brain, even at birth, that is essential for the full range of human grammatical expression.

When language has been displaced into the right hemisphere by a missing left hemisphere, visual-spatial abilities also suffer, usually more than language. This, in a sense, is the reverse of the situation in dyslexia: in the dyslexic child, the abnormal presence of visual-spatial functions in the left brain impairs language; in these right-brain-only children, the abnormal presence of language impairs visual-spatial functions. Perhaps it requires more brain than is available in one hemisphere to do both things well.

Despite this ability of the other side of the brain to pick up the load in early life, the underlying basis for lateralization is probably wired into the human brain. A number of anatomical differences can be seen between left and right brain if one looks carefully enough. These asymmetries are present in the developing fetus, not just in the adult. The best-known of these asymmetries involves a region on the top of the temporal lobe (the planum temporale), which is about the size of a large coin on the left side but the size of a small coin on the right side. This asymmetry is reversed (right side bigger than left) in just about the proportion of people that one would expect not to have language in the left brain. Like lateralization of visual-spatial function to the right brain, this asymmetry is greater in men than women.

If this anatomical asymmetry is concomitant with the lateralization of language, might other animals also have it? Might it mark some of the final evolutionary steps toward language? Such an asymmetry has been identified in chimpanzee and orangutan, although their asymmetries are far less extensive (and thus harder to spot) than those of man. Chimpanzees have been taught simple language (using sign language, since chimps lack the requisite vocal development for versatile speech).

The right brain is concerned with visual-spatial functions. Strokes that damage the right brain may leave the patient looking remarkably normal except, perhaps, for left-body paralysis if the stroke is large enough to also damage the right motor strip. But the fluent speech of such stroke patients conceals a severe deficit, a deficit that could have major consequences. Imagine the following drama:

The President of the United States is lying in bed, waving his right hand at his Secretary of State in a gesture of dismissal. The President is alert and seems intelligent. He is talking forcefully, angry at his subordinate, who has suggested that the President is ill and perhaps should delegate some of his duties to others until he recovers.

Indeed, the President's left side seems to be totally paralyzed -from a recent stroke. His left arm lies limp. The President cannot walk because his left leg will not function. However, the President seems blissfully unaware of this disorder, steadfastly denying that there is anything wrong with him. It is, of course, this denial of his illness that has particularly upset the President's personal and official families. They have tried to reason with him, pointing out to him that his left arm is lying there, paralyzed. But he denies that it is his left arm. Indeed, he is somewhat puzzled about what a strange arm and leg are doing in his bed with him.

President Woodrow Wilson suffered a stroke in 1919, after he triumphantly negotiated the League of Nations charter but before he attempted to persuade the U.S. Senate to ratify it. The stroke paralyzed his left side. He denied his illness to the point of paralyzing his administration; he fired his Secretary of State for discussing the illness with the Cabinet. Such effects of right-brain strokes were only beginning to be recognized by neurologists of that era. Today, the "denial of illness" syndrome is well known, and the typical symptoms of its victims have been ascribed to our president of the above paragraph. It illustrates how the right parietal lobe plays a role in the perception of one's own body image; if one cannot perceive anything about the left side of the body, to say nothing is wrong with it does exhibit a certain logic.

Less severe strokes provide another illustration of parietal lobe function. In a small right-brain stroke confined to the parietal lobe, there may be no paralysis, no visual defects, no other sensory defects of the usual sorts. Indeed, the person appears normal. But if the patient is driving a car approaching an intersection, though he can see cars approaching from the left just as easily as he can see cars coming from the right, if cars approach from both directions at the same time, the patient is likely to pay attention only to the one coming from the right. He may never realize that the left one exists, unless, of course, a collision happens to occur. If touched on both hands at the same time, he may notice only the touch to the right hand.

It may be very difficult to persuade such a patient that he should not drive a car. He may not be able to tell that anything is wrong with himself. He cannot perceive his own deficit in perception. It is this loss of insight into one's own abilities that is a particularly devastating feature of these strokes; these are among the most difficult brain-damaged patients to rehabilitate. Such unawareness of a stroke is an exception to the general rule. Disorders of the brain are truly dehumanizing conditions, made all the more unbearable because of the patient's often agonizing awareness that he or she is no longer the same person.

Damage to the right parietal lobe can also produce deficits in other visual-spatial abilities. The patient may have difficulty reading a map, or finding the way from bed to bathroom. He may have difficulty dressing himself. He can move all of his muscles in a coordinated manner but his arm may wind up in a pants leg rather than in a sleeve, even though he can name sleeves and pant legs and describe what they are used for. As with our President, such patients seem to have a disturbance in their internal image of their own bodies.

If the patient is an artist, his drawings may be distorted, perhaps only filling in one side of the canvas, or only developing one side of a face. Neurologists regularly test forms of artistic skills as a way of detecting right-parietal-lobe damage. Such patients produce defective drawings, even if only copying a cross or a house or a clock face. Features are omitted and others are crowded over onto the right side of the drawing. A clock face may have all twelve numbers, but crammed into the right side between twelve and six o'clock; the left-side positions between seven and eleven are ignored. The patients seem to realize that clocks should have twelve numbers but cannot space them out around the left half of the circle.

These deficits not only affect the performance of an artist, but many other skills as well. The auto mechanic can identify and explain the function of all the parts of an engine after such brain damage, but is unable to assemble the parts. It has been suggested that unusually good functioning of this nondominant (usually right) parietal lobe is a feature of those who excel in the visual arts: painters, sculptors, architects, movie makers.

Other skills seem to depend on function of both sides of the brain. Mathematical calculations are often impaired by damage to either parietal lobe, with division and subtraction more readily impaired than addition or multiplication. Dichhaptic testing suggests that the right brain is more important than the left for this calculating ability. The defect with left parietal damage may, in part, be difficulty in labeling mathematical symbols with words, for it is sometimes associated with several other symptoms that involve putting word labels on "spatial" material: inability to distinguish left and right and to name body parts (particularly, and sometimes only, the inability to name the fingers). Writing is also frequently impaired, as is spelling; yet, in some such patients, other language abilities are largely intact, such as naming or reading.

Drawings following a small right-parietal-lobe stroke, damaging the area shown in the right-brain map.

1. A cross drawn by the examiner for the patient to copy

2. The patient's attempt at copying the cross

3. The examiner's drawing of a house

4. The patient's attempt at copying it; again, the features on the left side are omitted

5. The examiner drew a circle, added some hair, and asked the patient to draw a face inside the circle; the patient drew features only on the right side of the circle (he said that the tongue was sticking out).

6. The examiner drew the circle indicated by the arrow and asked the patient to draw a sunflower; the patient filled in the petals and leaves

After all of the drawings were complete, the examiner asked the patient if anything was wrong with any of the drawings; the patient said no. The patient could name the object in each drawing, could name the various parts of the face and sunflower, and was apparently quite oblivious to the missing features in the left halves.

Musical abilities also seem to be distributed on both sides of brain. The right brain plays a major role in music, for during the time that only the right brain is put to sleep by barbiturate injections, the patient sings in a monotone, having lost the ability to correctly reproduce pitch; rhythm is less disturbed. Temporal lobe seems to be especially important to music. Right-temporal-lobe damage disturbs memory for tone sequences and for loudness, though rhythm memory is relatively unaffected. Left-temporal-lobe damage does not alter memory for tone or loudness, though it may affect rhythm. Yet damage to parts of the left temporal lobe will sometimes greatly interfere with musical abilities. Emotional responses to music are said to be diminished after left-brain damage. It has been suggested that skilled musicians develop a left-brain dominance for music compared to the right-brain dominance of the average person.

The areas of the left brain involved in music are separate from those for language, however, as illustrated by the following two patients. Because of this, music therapy has been useful in the rehabilitation of some aphasic patients, especially those with deficits in language fluency.

Ron and Norman were admitted on the same day with blood clots in their left brains. Ron's hobby was singing folk songs. After recovering from the acute effect of his hemorrhage, he found that he couldn't sing at all. Neither words nor tune came, even for his old favorites. His speech was pretty good, though, just an occasional name that wouldn't come to mind. In contrast, Norman's speech was terrible, full of jargon, and sometimes he couldn't follow verbal commands. But a speech therapist finally got across the idea of singing to Norman, and he sang very well, including all kinds of words that he couldn't speak correctly. CT scan showed Ron's and Norman's hemorrhages to be about the same size, but Ron's was in the front part of the left temporal lobe, Norman's in the back part.

As far as the brain is concerned, there seems to be a difference between speaking and singing the same string of words. Does the tune provide a framework which helps Norman put the words in sequence? Does Ron's inability to sing reflect a failure in the coordination between word strings and the tone sequence? We would like answers to such speculative questions, but how can we get them? In science, one tries to phrase the question (i.e., design the experiment) in such a way as to force nature to give an unambiguous answer. Strokes are intrinsically messy experiments performed by chance. Electrical stimulation, as we have seen, is more localizable and controllable (and, most important, reversible!).

But the brain of an epileptic may not be normal (in the sense that many epileptics have suffered seizures since childhood, and the brain's flexibility during development might rearrange maps); indeed, this is a major problem in interpreting split-brain studies to infer the capabilities of the right brain. Another important aspect of experimental design is what one asks the patient to do. One must accurately measure a particular, limited kind of behavior (such as naming, post-distractional memory for words, phoneme recognition, etc.). If one wanted to design experiments around our speculative questions relating word sequence to tone sequence, one would have to come up with some ingenious verbal tests and then find some technological method of measuring or disrupting neuronal function.

Thus, the "maps" of cortical function may be biased by the techniques available and by the questions asked. But there is also another way of finding where different brain functions are located. When neurons work, they increase the blood flow locally. So if blood flow is measured during a test of specific behavior such as reading, the areas of brain that are working harder during that behavior can be detected. " Such blood-flow measurements can be made through the intact head by introducing a radioactive tracer in the blood (radioactive xenon is commonly used) and placing a series of radioactivity counters over the scalp. The counters are designed to measure just the radioactivity coming from brain directly beneath them. The rate at which the tracer is cleared from beneath the counters is determined, a measurement proportional to blood flow. At present this technique cannot adequately discriminate between blood flow in cortex and that in underlying brain. But such measurements confirm, in the intact person, many of the findings derived from the study of stroke patients or from stimulation mapping of epileptics.

Each of the techniques has its limitations, but together they give us a picture of cortical localization. The blood-flow techniques have also allowed studies to be made of the cortical areas active in mentally rehearsing an act without making any movements. In silent reading, parts of the frontal lobe become quite active, together with language and visual areas; when reading aloud, the motor strip's oral and facial areas also increase their activity. This may suggest an orchestra with the conductor bringing in the trumpets to add to the theme developed by the strings and the woodwinds. But to what extent is this analogy valid, or is the orchestra conductor really just another version of the "little-man-in-the-head" reasoning which has plagued out thinking about the brain? To explore the topic, we will consider the brain's mechanisms controlling attention and arousal, initially by illustrating some of their malfunctions.