Right Brain of Language - University of Phoenix

16Lateralization, Language,

and the Split BrainThe Left Brain and the Right Brain of Language

16.1 Cerebral Lateralization of

Function: Introduction

16.2 The Split Brain

16.3 Differences between the Left

and Right Hemispheres

16.4 Cortical Localization of Language:

The Wernicke-Geschwind Model

16.5 Evaluation of the Wernicke-Geschwind Model

16.6 Cognitive Neuroscience Approach

to Language

16.7 Cognitive Neuroscience Approach

and Dyslexia

ISB

N0-558-84818-4

Biopsychology, Seventh Edition, by John P.J. Pinel. Published by Allyn & Bacon. Copyright © 2009 by Pearson Education, Inc.

40116.1 ■ Cerebral Lateralization of Function: Introduction

With the exception of a few midline orifices, wehumans have two of almost everything—one onthe left and one on the right. Even the brain,

which most people view as the unitary and indivisiblebasis of self, reflects this general principle of bilateral du-plication. In its upper reaches, the brain comprises twostructures, the left and right cerebral hemispheres, whichare entirely separate except for the cerebral commissuresconnecting them. The fundamental duality of the humanforebrain and the locations of the cerebral commissuresare illustrated in Figure 16.1.

Although the left and right hemispheres are similar inappearance, there are major differences between them in function. This chapter is about these differences, a topic commonly referred to as lateralization of func-tion. The study of split-brain patients—patients whose left and right hemispheres have been separated bycommissurotomy—is a major focus of discussion. An-other focus is the cortical localization of language abilitiesin the left hemisphere; language abilities are the mosthighly lateralized of all cognitive abilities.

You will learn in this chapter that your left and righthemispheres have different abilities and that they have thecapacity to function independently—to have differentthoughts, memories, and emotions. Accordingly, thischapter will challenge the concept you have of yourself asa unitary being. I hope you both enjoy it.

16.1Cerebral Lateralization ofFunction: Introduction

In 1836, Marc Dax, an unknown country doctor, pre-sented a short report at a medical society meeting inFrance. It was his first and only scientific presentation.Dax was struck by the fact that of the 40 or so brain-damaged patients with speech problems whom he hadseen during his career, not a single one had damage re-stricted to the right hemisphere. His report aroused littleinterest, and Dax died the following year unaware that he

Hippocampalcommissure

Frontal section of the human brain, whichillustrates the fundamental duality of thehuman forebrain.

Midsagittal section of the human brain,which illustrates the corpus callosum andother commissures.

Corpuscallosum

Massaintermedia

Anteriorcommissure

Opticchiasm

Posteriorcommissure

FIGURE 16.1 The cerebral hemispheres and cerebral commissures.

ISB

N0-

558-

8481

8-4


had anticipated one of the most important areas of mod-ern neuropsychological research.

Discovery of the Specific Contributions of Left-Hemisphere Damage to Aphasia and Apraxia

One reason Dax’s important paper had so little impactwas that most of his contemporaries believed that thebrain acted as a whole and that specific functions couldnot be attributed to particular parts of it. This view beganto change 25 years later, when Paul Broca reported his postmortem examination of two aphasic patients.Aphasia is a brain-damage-produced deficit in the abilityto produce or comprehend language.

Both of Broca’s patients had a left-hemisphere lesionthat involved an area in the frontal cortex just in front of

the face area of the primary motor cor-tex. Broca at first did not realize thatthere was a relation between aphasia

and the side of the brain damage; he had not heard ofDax’s report. However, by 1864, Broca had performedpostmortem examinations on seven more aphasic pa-tients, and he was struck by the fact that, like his first two,they all had damage to the inferior prefrontal cortex of theleft hemisphere—which by then had become known asBroca’s area (see Figure 16.2).

In the early 1900s, another example of cerebral lateral-ization of function was discovered. Hugo-Karl Liepmannfound that apraxia, like aphasia, is almost always associ-ated with left-hemisphere damage, despite the fact that itssymptoms are bilateral (involving both sides of the body).Apraxic patients have difficulty performing movementswhen asked to perform them out of context, even thoughthey often have no difficulty performing the same move-ments when they are not thinking about doing so.

The combined impact of the evidence that the lefthemisphere plays a special role in both language and voluntary movement led to the conceptof cerebral dominance. According to thisconcept, one hemisphere—usually theleft—assumes the dominant role in the control of allcomplex behavioral and cognitive processes, and theother plays only a minor role. This concept led to thepractice of referring to the left hemisphere as the dominant hemisphere and the right hemisphere asthe minor hemisphere.

Tests of Cerebral Lateralization

Early research on the cerebral lateralization of functioncompared the effects of left-hemisphere and right-hemisphere lesions. Now, however, other techniques arealso used for this purpose. The sodium amytal test, the di-chotic listening test, and functional brain imaging arethree of them.

Sodium Amytal Test The sodium amytal test of lan-guage lateralization (Wada, 1949) is often given to patientsprior to neurosurgery. The neurosurgeon uses the resultsof the test to plan the surgery; every effort is made to avoiddamaging areas of the cortex that are likely to be involvedin language. The sodium amytal test involves the injectionof a small amount of sodium amytal into the carotid ar-tery on one side of the neck. The injection anesthetizes thehemisphere on that side for a few minutes, thus allowingthe capacities of the other hemisphere to be assessed. Dur-ing the test, the patient is asked to recite well-known series(e.g., letters of the alphabet, days of the week, months ofthe year) and to name pictures of common objects. Then,an injection is administered to the other side, and the testis repeated. When the hemisphere that is specialized forspeech, usually the left hemisphere, is anesthetized, the pa-tient is rendered completely mute for a minute or two;then once the ability to talk returns, there are errors of se-rial order and naming. In contrast, when the minor speechhemisphere, usually the right, is anesthetized, mutismoften does not occur at all, and errors are few.

Dichotic Listening Test Unlike the sodium amytaltest, the dichotic listening test is noninvasive; thus, it canbe administered to healthy subjects. In the standard di-chotic listening test (Kimura, 1961), three pairs of spoken

402 Chapter 16 ■ Lateralization, Language, and the Split Brain

Primary motorcortexBroca’s

area Parietal lobe

Centralfissure

OccipitallobeTemporal

lobe

Lateralfissure

Frontallobe

ClinicalClinicalImplicationsImplications

FIGURE 16.2 The location of Broca’s area: in the inferior leftprefrontal cortex, just anterior to the face area of the left pri-mary motor cortex.


ISB

N0-558-84818-4


digits are presented through earphones; the digits of eachpair are presented simultaneously, one to each ear. For ex-ample, a subject might hear the sequence 3, 9, 2 throughone ear and at the same time 1, 6, 4 through the other. Thesubject is then asked to report all of the digits. Kimurafound that most people report slightly more of the digitspresented to the right ear than the left, which is indicativeof left-hemisphere specialization for language. In con-trast, Kimura found that all the patients who had beenidentified by the sodium amytal test as having right-hemisphere specialization for language performed betterwith the left ear than the right.

Why does the superior ear on the dichotic listening testindicate the language specialization of the contralateralhemisphere? Kimura argued that although the soundsfrom each ear are projected to both hemispheres, the con-tralateral connections are stronger and take precedencewhen two different sounds are simultaneously competingfor access to the same cortical auditory centers.

Functional Brain Imaging Lateralization of functionhas also been studied using functional brain-imagingtechniques. While the subject engages in some activity,such as reading, the activity of the brain is monitored bypositron emission tomography (PET) or functional mag-netic resonance imaging (fMRI). On language tests, func-tional brain-imaging techniques typically reveal fargreater activity in the left hemisphere than in the righthemisphere (see Martin, 2003).

Discovery of the Relation between SpeechLaterality and Handedness

Two early large-scale lesion studies clarified the relationbetween the cerebral lateralization of speech and handed-ness. One study involved military personnel who suffered

brain damage in World War II (Russell& Espir, 1961), and the other focused onneurological patients who underwent

unilateral excisions for the treatment of neurological dis-orders (Penfield & Roberts, 1959). In both studies, ap-proximately 60% of dextrals (right-handers) withleft-hemisphere lesions and 2% of those with right-hemisphere lesions were diagnosed as aphasic; the com-parable figures for sinestrals (left-handers) were about30% and 24%, respectively. These results indicate that theleft hemisphere is dominant for language-related abilitiesin almost all dextrals and in the majority of sinestrals.Consequently, sinestrals are more variable (less pre-dictable) than dextrals with respect to their hemisphere oflanguage lateralization.

Results of the sodium amytal test have confirmed therelation between handedness and language lateralization

that was first observed in early lesionstudies. For example, Milner (1974)found that almost all right-handed pa-

tients without early left-hemisphere damage had left-hemisphere specialization for speech (92%), that mostleft-handed and ambidextrous patients without early left-hemisphere damage had left-hemisphere specializationfor speech (69%), and that early left-hemisphere damagedecreased left-hemisphere specialization for speech inleft-handed and ambidextrous patients (30%).

In interpreting Milner’s figures, it is important to re-member that sodium amytal tests are administered onlyto people who are experiencing braindysfunction, that early brain damagecan cause the lateralization of speech toshift to the other hemisphere (see Maratsos & Matheny,1994; Stiles, 1998), and that many more people have left-hemisphere specialization for speech to start with. Con-sidered together, these points suggest that Milner’sfindings likely underestimate the proportion of peoplewith left-hemisphere specialization for speech amonghealthy members of the general population.

Sex Differences in Brain Lateralization

Interest in the possibility that the brains of females andmales differ in their degree of lateralization was stimu-lated by McGlone’s (1977, 1980) studies of unilateralstroke victims. McGlone found thatmale victims of unilateral strokes werethree times more likely to suffer fromaphasia than female victims. She found that male victimsof left-hemisphere strokes had deficits on the WechslerAdult Intelligence Scale (WAIS) verbal subtests, whereasmale victims of right-hemisphere strokes had deficits onthe WAIS performance subtests. In contrast, in femalevictims, there were no significant differences between thedisruptive effects of left- and right-hemisphere unilateralstrokes on performance on the WAIS. On the basis ofthese three findings, McGlone concluded that the brainsof males are more lateralized than the brains of females.

McGlone’s hypothesis of a sex difference in brain later-alization has been widely embraced, and it has been usedto explain almost every imaginable behavioral differencebetween the sexes. But support for McGlone’s hypothesishas been mixed. Some researchers have failed to confirmher report of a sex difference in the effects of unilateralbrain lesions (see Inglis & Lawson, 1982). In addition, al-though a few functional brain-imaging studies have sug-gested that females, more than males, use bothhemispheres in the performance of language-related tasks(e.g., Jaeger et al., 1998; Kansaku, Yamaura, & Kitazawa,2000), a meta-analysis of 14 functional brain-imagingstudies did not find a significant effect of sex on languagelateralization (Sommer et al., 2004).

So far, you have learned about four methods of study-ing cerebral lateralization of function: comparing the ef-fects of unilateral left- and right-hemisphere brainlesions, the sodium amytal test, the dichotic listening test,

40316.1 ■ Cerebral Lateralization of Function: Introduction

ClinicalClinicalImplications


NeNeuroplasticityroplasticity


ISB

N0-

558-

8481

8-4



and functional brain imaging. The next section describesa fifth method.

16.2The Split Brain

In the early 1950s, the corpus callosum—the largest cere-bral commissure—constituted a paradox of major pro-portions. Its size, an estimated 200 million axons, and itscentral position, right between the two cerebral hemi-spheres, implied that it performed an extremely impor-tant function; yet research in the 1930s and 1940s seemedto suggest that it did nothing at all. The corpus callosumhad been cut in monkeys and in several other laboratoryspecies, but the animals seemed no different after the sur-gery than they had been before. Similarly, human patientswho were born without a corpus callosum or had it dam-aged seemed perfectly normal. In the early 1950s, RogerSperry, whom you may remember from the eye-rotationexperiments of Chapter 9, and his colleagues were in-trigued by this paradox.

Groundbreaking Experiment of Myers and Sperry

The solution to the puzzle of the corpus callosum wasprovided in 1953 by an experiment on cats by Myers and

Sperry. The experiment made two as-tounding theoretical points. First, itshowed that one function of the corpus

callosum is to transfer learnedinformation from one hemi-sphere to the other. Second, itshowed that when the corpus cal-losum is cut, each hemispherecan function independently;each split-brain cat appeared tohave two brains. If you find thethought of a cat with two brainsprovocative, you will almost cer-tainly be bowled over by similarobservations about split-brain

humans. But I am getting ahead of myself. Let’s first con-sider the research on cats.

In their experiment, Myers and Sperry trained cats toperform a simple visual discrimination. On each trial,each cat was confronted by two panels, one with a circleon it and one with a square on it. The relative positions ofthe circle and square (right or left) were varied randomlyfrom trial to trial, and the cats had to learn which symbolto press in order to get a food reward. Myers and Sperrycorrectly surmised that the key to split-brain research wasto develop procedures for teaching and testing one hemi-sphere at a time. Figure 16.3 illustrates the method theyused to isolate visual-discrimination learning in onehemisphere of the cats. There are two routes by which vi-sual information can cross from one eye to the contralat-eral hemisphere: via the corpus callosum or via the opticchiasm. Accordingly, in their key experimental group,Myers and Sperry transected (cut completely through)both the optic chiasm and the corpus callosum of each catand put a patch on one eye. This restricted all incomingvisual information to the hemisphere ipsilateral to theuncovered eye.

The results of Myers and Sperry’s experiment are illus-trated in Figure 16.4. In the first phase of the study, all catslearned the task with a patch on one eye. The cats in thekey experimental group (those with both the optic chi-asm and the corpus callosum transected) learned the sim-ple discrimination as rapidly as did unlesioned controlcats or control cats with either the corpus callosum or the optic chiasm transected, despite the fact that cuttingthe optic chiasm produced a scotoma—an area ofblindness—involving the entire medial half of each

The Evolutionahe Evolutionary PerspectivePerspective

Transectedcorpuscallosum

Blindfoldedone eye

Transectedopticchiasm

FIGURE 16.3 Restricting visualinformation to one hemisphere incats. To restrict visual information toone hemisphere, Myers and Sperry(1) cut the corpus callosum, (2) cutthe optic chiasm, and (3) blindfoldedone eye. This restricted the visual in-formation to the hemisphere ipsi-lateral to the uncovered eye.

ISB

N0-558-84818-4


40516.2 ■ The Split Brain

retina. This result suggested that one hemisphere workingalone can learn simple tasks as rapidly as two hemispheresworking together.

More surprising were the results of the second phase ofMyers and Sperry’s experiment, during which the patchwas transferred to each cat’s other eye. The transfer of thepatch had no effect on the performance of the intact con-trol cats or of the control cats with either the optic chiasmor the corpus callosum transected; these subjects contin-ued to perform the task with close to 100% accuracy. Incontrast, transferring the eye patch had a devastating ef-fect on the performance of the experimental cats. In ef-fect, it blindfolded the hemisphere that had originallylearned the task and tested the knowledge of the otherhemisphere, which had been blindfolded during initial

training. When the patch was transferred, the perform-ance of the experimental cats dropped immediately tobaseline (i.e., to 50% correct); and then the cats relearnedthe task with no savings whatsoever, as if they had neverseen it before. Myers and Sperry concluded that the catbrain has the capacity to act as two separate brains andthat the function of the corpus callosum is to transmit in-formation between them.

Myers and Sperry’s startling conclusions about thefundamental duality of the cat brain and the informa-tion-transfer function of the corpus cal-losum have been confirmed in a varietyof species with a variety of test proce-dures. For example, split-brain monkeys cannot performtasks requiring fine tactual discriminations (e.g., rough


Experimental group

Control groups

Cor

rect

cho

ices

(

or

)

Patch on first eye Patch on second eye

Trials50%

100%

Cats with either their optic chiasm transected, corpus callosum transected, or neither transected (shown here) learned the discrimination at a normal rate with one eye blindfolded and retained the task perfectly when the blindfold was switched to the other eye.

Cor

rect

cho

ices

(

or

)

Patch on first eye Patch on second eye

Trials50%

100%

Cats with both their optic chiasms and corpus callosums transected learned the discrimination at a normal rate with one eye blindfolded, but they showed no retention whatsoever when the blindfold was switched to the other eye.

FIGURE 16.4 Schematic illustration of Myers and Sperry’s (1953) groundbreaking split-brain ex-periment. There were four groups: (1) the key experimental group with both the optic chiasm andcorpus callosum transected, (2) a control group with only the optic chiasm transected, (3) a controlgroup with only the corpus callosum transected, and (4) an unlesioned control group. The per-formance of the three control groups did not differ, so they are illustrated here together.

ISB

N0-

558-

8481

8-4



versus smooth) or fine motor responses (e.g., unlocking apuzzle) with one hand if they have learned them with theother—provided that they are not allowed to watch theirhands, which would allow the information to enter bothhemispheres. There is no transfer of fine tactual andmotor information in split-brain monkeys because thesomatosensory and motor fibers involved in fine sensoryand motor discriminations are all contralateral.

Commissurotomy in Human Epileptics

In the first half of the 20th century, when the normal func-tion of the corpus callosum was still a mystery, it wasknown that epileptic discharges often spread from onehemisphere to the other through the corpus callosum.This fact, along with the fact that cutting the corpus callo-sum had proven in numerous studies to have no obviouseffect on performance outside the contrived conditions ofSperry’s laboratory, led two neurosurgeons, Vogel andBogen, to initiate a program of commissurotomy for thetreatment of severe intractable cases of epilepsy—despitethe fact that a previous similar attempt had failed, pre-sumably because of incomplete transections (Van Wage-nen & Herren, 1940).

The rationale underlying therapeuticcommissurotomy—which typically in-

volves transecting the corpus callosum and leaving thesmaller commissures intact—was that the severity of thepatient’s convulsions might be reduced if the dischargescould be limited to the hemisphere of their origin. Thetherapeutic benefits of commissurotomy turned out to beeven greater than anticipated: Despite the fact that com-missurotomy is performed in only the most severe cases,many commissurotomized patients do not experience an-other major convulsion.

The evaluation of the split-brain patient’s neuropsy-chological status was placed in the capable hands ofSperry and his associate Gazzaniga (the ensuing researchwas a major factor in Sperry’s being awarded a NobelPrize in 1981; see Table 1.1.) They began by developinga battery of tests based on the same methodological strat-egy that had proved so informative in their studies of laboratory animals: delivering information to onehemisphere while keeping it out of the other (see Gaz-zaniga, 2005).

They could not use the same visual-discriminationprocedure that had been used in studies of split-brain lab-oratory animals (i.e., cutting the optic chiasm and blind-folding one eye) because cutting the optic chiasmproduces a scotoma. Instead, they employed the proce-dure illustrated in Figure 16.5. Each patient was asked tofixate on the center of a display screen; then, visual stim-


APPLE SPOON

FIGURE 16.5 The testing procedure that was used to evaluate the neuropsychological status ofsplit-brain patients. Visual input goes from each visual field to the contralateral hemisphere; finetactile input goes from each hand to the contralateral hemisphere; and each hemisphere controlsthe fine motor movements of the contralateral hand.

ISB

N0-558-84818-4



uli were flashed onto the left or right side of the screen for0.1 second. The 0.1-second exposure time was longenough for the subjects to perceive the stimuli but shortenough to preclude the confounding effects of eye move-ment. All stimuli thus presented in the left visual fieldwere transmitted to the right visual cortex, and all stimulithus presented in the right visual field were transmitted tothe left visual cortex.

Fine tactual and motor tasks were performed by eachhand under a ledge. This procedure was used so that thenonperforming hemisphere, that is, the ipsilateral hemi-sphere, could not monitor the performance via the vi-sual system.

The results of the tests on split-brain patients haveconfirmed the findings in split-brain laboratory animalsin one major respect, but not in another. Like split-brainlaboratory animals, human split-brain patients seem tohave in some respects two independent brains, each withits own stream of consciousness, abilities, memories, andemotions (e.g., Gazzaniga, 1967; Gazzaniga & Sperry,1967; Sperry, 1964). But unlike the hemispheres of split-brain laboratory animals, the hemispheres of split-brainpatients are far from equal in their ability to perform cer-tain tasks. Most notably, the left hemisphere of most split-brain patients is capable of speech, whereas the righthemisphere is not.

Before I recount some of the key results of the testson split-brain humans, let me give you some advice. Somestudents become confused by the results of these testsbecause their tendency to think of the human brain asa single unitary organ is deeply engrained. If you be-come confused, think of each split-brain patient as twoseparate subjects: Ms. or Mr. Right Hemisphere, who un-derstands a few simple instructions but cannot speak,who receives sensory information from the left visualfield and left hand, and who controls the fine motor re-sponses of the left hand; and Ms. or Mr. Left Hemisphere,

who is verbally adept, who receives sen-sory information from the right visualfield and right hand, and who controls

the fine motor responses of the right hand. In everydaylife, the behavior of split-brain subjects is reasonably nor-mal because their two brains go through life togetherand acquire much of the same information; however, inthe neuropsychological laboratory, major discrepanciesin what the two hemispheres learn can be created. Asyou are about to find out, this situation has some inter-esting consequences.

Evidence That the Hemispheres of Split-Brain Patients Can Function Independently

If a picture of an apple were flashed in the right visualfield of a split-brain patient, the left hemisphere coulddo one of two things to indicate that it had received and stored the information. Because it is the hemisphere

that speaks, the left hemisphere could simply tell the experimenter that it saw a picture of an apple. Or thepatient could reach under the ledge with the right hand,feel the test objects that are there, and pick out the apple.Similarly, if the apple were presented to the left hemi-sphere by being placed in the patient’s right hand, theleft hemisphere could indicate to the experimenter thatit was an apple either by saying so or by putting the apple down and picking out another apple with theright hand from the test objects under the ledge. If,however, the nonspeaking right hemisphere were askedto indicate the identity of an object that had previouslybeen presented to the left hemisphere, it could not do so. Although objects that have been presented to the left hemisphere can be accurately identified with theright hand, performance is no better than chance withthe left hand.

When test objects are presented to the right hemi-sphere either visually (in the left visual field) or tactu-ally (in the left hand), the pattern of responses is entirelydifferent. A split-brain patient asked to name an objectflashed in the left visual field is likely to claim that noth-ing appeared on the screen. (Remember that it is the lefthemisphere who is talking and the right hemisphere whohas seen the stimulus.) A patient asked to name an ob-ject placed in the left hand is usually aware that some-thing is there, presumably because of the crude tactualinformation carried by ipsilateral somatosensory fibers,but is unable to say what it is (see Fabri et al., 2001).Amazingly, all the while the patient is claiming (i.e., allthe while the left hemisphere is claiming) the inabilityto identify a test object presented in the left visual fieldor left hand, the left hand (i.e., the right hemisphere)can identify the correct object. Imagine how confusedthe patient must become when, in trial after trial, the lefthand can feel an object and then fetch another just likeit from a collection of test items under the ledge, whilethe left hemisphere is vehemently claiming that it doesnot know the identity of the test object.

Cross-Cuing

Although the two hemispheres of a split-brain subjecthave no means of direct neural communication, they cancommunicate neurally via indirect pathways through thebrain stem. They can also communicate with each otherby an external route, by a process called cross-cuing. Anexample of cross-cuing occurred during a series of testsdesigned to determine whether the left hemisphere couldrespond to colors presented in the left visual field. Totest this possibility, a red or a green stimulus was pre-sented in the left visual field, and the split-brain subjectwas asked to verbally report the color: red or green. Atfirst, the patient performed at a chance level on this task(50% correct); but after a time, performance improvedappreciably, thus suggesting that the color information

Thinking ClearlyThinking Clearly

ISB

N0-

558-

8481

8-4



was somehow being transferred over neural pathwaysfrom the right hemisphere to the left. However, thisproved not to be the case:

We soon caught on to the strategy the patient used. If ared light was flashed and the patient by chance guessedred, he would stick with that answer. If the flashed lightwas red, and the patient by chance guessed green, hewould frown, shake his head and then say, “Oh no, Imeant red.” What was happening was that the right hemi-sphere saw the red light and heard the left hemispheremake the guess “green.” Knowing that the answer waswrong, the right hemisphere precipitated a frown and ashake of the head, which in turn cued in the left hemi-sphere to the fact that the answer was wrong and that ithad better correct itself! . . . The realization that the neu-rological patient has various strategies at his commandemphasizes how difficult it is to obtain a clear neurologi-cal description of a human being with brain damage.(Gazzaniga, 1967, p. 27)

Doing Two Things at Once

In most of the classes I teach, there is a student who fitsthe following stereotype. He sits—or rather sprawls—near the back of the class; and despite good grades, hetries to create the impression that he is above it all bymaking sarcastic comments. I am sure you recognizehim—and it is almost always a him. Such a student inad-vertently triggered an interesting discussion in one of myclasses. His comment went something like this: “If gettingmy brain cut in two could create two separate brains, per-haps I should get it done so that I could study for two dif-ferent exams at the same time.”

The question raised by this comment is a good one. Ifthe two hemispheres of a split-brain patient are capable ofindependent functioning, then they should be able to dotwo different things at the same time—in this case, learntwo different things at the same time. Can they? Indeedthey can. For example, in one test, two different visualstimuli appeared simultaneously on the test screen—let’ssay a pencil in the left visual field and an orange in theright visual field. The split-brain patient was asked to si-multaneously reach into two bags—one with each hand—and grasp in each hand the object that was on the screen.After grasping the objects, but before withdrawing them,the subject was asked to tell the experimenter what was inthe two hands; the subject (i.e., the left hemisphere)replied, “Two oranges.” Much to the bewilderment of theverbal left hemisphere, when the hands were withdrawn,there was an orange in the right hand and a pencil in theleft. The two hemispheres of the split-brain subject hadlearned two different things at exactly the same time.

In another test in which two visual stimuli were pre-sented simultaneously—again, let’s say a pencil to the leftvisual field and an orange to the right—the split-brainsubject was asked to pick up the presented object from an

assortment of objects on a table, this time in full view. Asthe right hand reached out to pick up the orange underthe direction of the left hemisphere, the right hemispheresaw what was happening and thought an error was beingmade (remember that the right hemisphere saw a pencil).On some trials, the right hemisphere dealt with this prob-lem in the only way that it could: The left hand shot out,grabbed the right hand away from the orange, and redi-rected it to the pencil. This response is called the helping-hand phenomenon.

The special ability of split brains to do two things atonce has also been demonstrated on tests of attention.Each hemisphere of split-brain patients appears to be ableto maintain an independent focus of attention (see Gaz-zaniga, 2005). This leads to an ironic pattern of results:Split-brain patients can search for, and identify, a visualtarget item in an array of similar items more quickly thana healthy control subject (Luck et al., 1989)—presumablybecause the two split hemispheres are conducting two in-dependent searches.

Yet another example of the split-brain’s special abil-ity to do two things at once involves the phenomenonof visual completion. As you may recall from Chapter6, individuals with scotomas are often unaware of thembecause their brains have the capacity to fill them in (tocomplete them) by using information from the surround-ing areas of the visual field. In a sense, each hemisphereof a split-brain patient is a subject with a scotoma cov-ering the entire ipsilateral visual field. The ability of thehemispheres of a split-brain patient to simultaneouslyand independently engage in completion has beendemonstrated in studies using the chimeric figurestest—named after Chimera, a mythical monster com-posed of the combined parts of different animals. Levy,Trevarthen, and Sperry (1972) flashed photographs com-posed of fused-together half-faces of two different peo-ple onto the center of a screen in front of their split-brainsubjects. The subjects were then asked to describe whatthey saw or to indicate what they saw by pointing to itin a series of photographs of intact faces. Amazingly, eachsubject (i.e., each left hemisphere) reported seeing a com-plete, bilaterally symmetrical face, even when asked suchleading questions as “Did you notice anything peculiarabout what you just saw?” When the subjects were askedto describe what they saw, they usually described a com-pleted version of the half that had been presented to theright visual field (i.e., the left hemisphere).

The Z Lens

Once it was firmly established that the two hemispheresof each split-brain patient can function independently, itbecame clear that the study of split-brain patients pro-vided a unique opportunity to compare the abilities ofleft and right hemispheres. However, early studies of the

ISB

N0-558-84818-4


Telescope projects the entire visual field onto the Z lens

The Z lens focuses the entire visual field on half the retina

One eye is covered

The entire visual field is projected to onehemisphere

FIGURE 16.6 The Z lens, which was developed by Zaidel tostudy functional asymmetry in split-brain patients. It is a contactlens that is opaque on one side (left or right), so that visual inputreaches only one hemisphere.


lateralization of function in split-brain patients were lim-ited by the fact that visual stimuli requiring more than 0.1second to perceive could not be studied using the conven-tional method for restricting visual input to one hemi-sphere. This methodological barrier was eliminated byZaidel in 1975. Zaidel developed a lens, called the Z lens,that limits visual input to one hemisphere of split-brainpatients while they scan complex visual material such asthe pages of a book. As Figure 16.6 illustrates, the Z lensis a contact lens that is opaque on one side (left or right).Because it moves with the eye, it permits visual input toenter only one hemisphere, irrespective of eye movement.Zaidel used the Z lens to compare the ability of the leftand right hemispheres of split-brain patients to performvarious tests.

The usefulness of the Z lens is not restricted to purelyvisual tests. For example, it has been used to compare the

ability of the left and right hemispheres to comprehendspeech. Because each ear projects to both hemispheres, itis not possible to present spoken words to only one hemi-sphere. Thus, to assess the ability of a hemisphere to com-prehend spoken words or sentences, Zaidel presentedthem to both ears, and then he asked the subject to pickthe correct answer or to perform the correct responseunder the direction of visual input to only that hemi-sphere. For example, to test the ability of the right hemi-sphere to understand oral commands, the subjects weregiven an oral instruction (such as “Put the green squareunder the red circle”), and then the right hemisphere’sability to comprehend the direction was tested by allowingonly the right hemisphere to observe the colored tokenswhile the task was being completed.

Dual Mental Functioning and Conflict in Split-Brain Patients

In most split-brain patients, the right hemisphere doesnot seem to have a strong will of its own; the left hemi-sphere seems to control most everyday activities. How-ever, in a few split-brain patients, the right hemispheretakes a more active role in controlling behavior, and inthese cases, there can be serious conflicts between the leftand right hemispheres. One patient (let’s call him Peter)was such a case.

The Case of Peter, the Split-BrainPatient Tormented by Conflict

At the age of 8, Peter began to suffer from complex partialseizures. Antiepileptic medication was ineffective, and at 20, he received a commissurotomy,which greatly improved his conditionbut did not completely block hisseizures. A sodium amytal test administered prior to surgery showed that he was left-hemisphere dominant for language.

Following surgery, Peter, unlike most other split-brainpatients, was not able to respond with the left side ofhis body to verbal input. When asked to make whole-bodymovements (e.g., “Stand like a boxer”) or movements of the left side of his body (e.g., “Touch your left ear with your left hand”), he could not respond correctly. Ap-parently, his left hemisphere could not, or would not,control the left side of his body via ipsilateral fibers. Dur-ing such tests, Peter—or, more specifically, Peter’s lefthemisphere—often remarked that he hated the left side ofhis body.

The independent, obstinate, and sometimes mischie-vous behavior of Peter’s right hemisphere often causedhim (his left hemisphere) considerable frustration. He


ISB

N0-

558-

8481

8-4



(his left hemisphere) complained that his left hand wouldturn off television shows that he was enjoying, that his leftleg would not always walk in the intended direction, andthat his left arm would sometimes perform embarrassing,socially unacceptable acts (e.g., striking a relative).

In the laboratory, he (his left hemisphere) sometimesbecame angry with his left hand, swearing at it, striking it,and trying to force it with his right hand to do what he(his left hemisphere) wanted. In these cases, his left handusually resisted his right hand and kept performing as di-rected by his right hemisphere. In these instances, it wasalways clear that the right hemisphere was behaving withintent and understanding and that the left hemispherehad no clue why the despised left hand was doing what itwas doing (Joseph, 1988).

Independence of Split Hemispheres: Current Perspective

Discussions of split-brain patients tend to focus on themany examples of complete functional independence, asI have done here. These examples are not only intriguingbut important, because they demonstrate major differ-ences from normal integrated brain function. However, itis important not to lose sight of the fact that surgicallyseparated hemispheres retain the ability to interact via thebrain stem and, in some cases, do function together. Thehemispheres of split-brain patients are more likely to per-form independently on some types of tests than on oth-ers, but sometimes two patients performing the same testmay differ in their hemispheric independence (see Wol-ford, Miller, & Gazzaniga, 2004).

The classic study of Sperry, Zaidel, and Zaidel (1979)provided early evidence that some types of informationare more likely to be shared between split hemispheres.These researchers used the Z lens to assess the behavioralreactions of the right hemispheres of split-brain patientsto various emotion-charged images: photographs of rela-tives; of pets; of themselves; and of political, historical,and religious figures and emblems. The subjects’ behav-ioral reactions were emotionally appropriate, thus indi-cating that right hemispheres are capable of emotionalexpression. In addition, there was an unexpected finding:The emotional content of images presented to the righthemisphere was reflected in the patients’ speech as well asin their nonverbal behavior. This suggested that emo-tional information was somehow being passed from theright to the verbal left hemisphere of the split-brain sub-jects. The ability of emotional reactions, but not visual in-formation, to be readily passed from the right hemisphereto the left hemisphere created a bizarre situation. A sub-ject’s left hemisphere often reacted with the appropriateemotional verbal response to an image that had been pre-sented to the right hemisphere, even though it did notknow what the image was.

Consider the following remarkable exchange (para-phrased from Sperry, Zaidel, & Zaidel, 1979, pp.161–162). The patient’s right hemisphere was presentedwith an array of photos, and the patient was asked if onewas familiar. He pointed to the photo of his aunt.

Experimenter: “Is this a neutral, a thumbs-up, or athumbs-down person?”

Patient: With a smile, he made a thumbs-up sign andsaid, “This is a happy person.”

Experimenter: “Do you know him personally?”Patient: “Oh, it’s not a him, it’s a her.”Experimenter: “Is she an entertainment personality or

an historical figure?”Patient: “No, just . . .”Experimenter: “Someone you know personally?”Patient: He traced something with his left index finger

on the back of his right hand, and then he ex-claimed, “My aunt, my Aunt Edie.”

Experimenter: “How do you know?”Patient: “By the E on the back of my hand.”

Another factor that has been shown to contribute sub-stantially to the hemispheric independence of split-brainpatients is task difficulty (Weissman & Banich, 2000). Astasks become more difficult, they are more likely to in-volve both hemispheres of split-brain patients. It appearsthat simple tasks are best processed in one hemisphere,the hemisphere specialized for the specific activity, butcomplex tasks require the cognitive power of both hemi-spheres. This is an important finding for two reasons.First, it complicates the interpretation of functionalbrain-imaging studies of lateralizationof function: If tasks are difficult, bothhemispheres may show substantial ac-tivity, even though one hemisphere is specialized for theperformance of the task. Second, it explains why the eld-erly often display less lateralization of function: As neuralresources decline, it may become necessary to involveboth hemispheres in most tasks.

16.3Differences between the Left and Right Hemispheres

So far in this chapter, you have learned about five meth-ods of studying cerebral lateralization of function: unilat-eral lesions, the sodium amytal test, the dichotic listeningtest, functional brain imaging, and studies of split-brainpatients. This section takes a look at some of the majorfunctional differences between the left and right cerebralhemispheres that have been discovered using these meth-ods. Because the verbal and motor abilities of the lefthemisphere are readily apparent (see Beeman &Chiarello, 1998; Reuter-Lorenz & Miller, 1998), most re-


ISB

N0-558-84818-4


41116.3 ■ Differences between the Left and Right Hemispheres

search on the lateralization of function has focused onuncovering the special abilities of the right hemisphere.

Slight Biases versus All-or-NoneHemispheric Differences

Before I introduce you to some of the differences betweenthe left and right hemispheres, I need to clear up a com-mon misconception: For many functions, there are nosubstantial differences between the hemispheres; andwhen functional differences do exist, these tend to beslight biases in favor of one hemisphere or the other—

not absolute differences (see Brown &Kosslyn, 1993). Disregarding thesefacts, the popular media inevitably por-

tray left–right cerebral differences as absolute. As a result,it is widely believed that various abilities reside exclusivelyin one hemisphere or the other. For example, it is widelybelieved that the left hemisphere has exclusive controlover language and the right hemisphere has exclusivecontrol over emotion and creativity.

Language-related abilities provide a particularly goodillustration of the fact that lateralization of function isstatistical rather than absolute. Language is the most lat-eralized of all cognitive abilities. Yet, even in this most ex-

treme case, lateralization is far from total; there is sub-stantial language-related activity in the right hemisphere.Following are three illustrations of this point: First, on thedichotic listening test, subjects who are left-hemispheredominant for language tend to identify more digits withthe right ear than the left ear, but this right-ear advantageis only slight, 55% to 45%. Second, in most split-brain pa-tients, the left hemisphere is dominant for language, butthe right hemisphere can understand many spoken orwritten words and simple sentences (see Baynes & Gaz-zaniga, 1997; Zaidel, 1987). And third, although there isconsiderable variability among split-brain patients intheir right-hemisphere performance on tests of languagecomprehension (Gazzaniga, 1998), the language abilitiesof their right hemispheres tend to be comparable to thoseof a preschool child.

Examples of Cerebral Lateralization of Function

The study of lateralization of function has put the archaicnotion of left-hemisphere dominance to rest. The righthemisphere has been shown to be functionally superior tothe left in many respects. Table 16.1 lists some of the abil-ities that have been shown to be lateralized. They are


TABLE 16.1 Abilities That Display Cerebral Lateralization of Function

GENERALFUNCTION

Right-HemisphereDominance

VISION

AUDITION

TOUCH

MOVEMENT

MEMORY

LANGUAGE

SPATIALABILITY

FacesGeometric patternsEmotional expression

WordsLetters

Nonlanguage soundsMusic

Tactile patternsBraille

Movement in spatial patterns

Nonverbal memoryPerceptual aspects of memories

Emotional content

Mental rotation of shapesGeometryDirectionDistance

Language sounds

Complex movementIpsilateral movement

Verbal memoryFinding meaning in memories

SpeechReading

WritingArithmetic

Left-HemisphereDominance

ISB

N0-

558-

8481

8-4



arranged in two columns: those that seem to be con-trolled more by the left hemisphere and those that seemto be controlled more by the right hemisphere.

This subsection describes several examples of cerebrallateralization of function. Because the chapter has so farfocused on the obvious specializations of the left hemi-sphere (i.e., on its verbal and motor specializations), thisdiscussion begins with the recent discovery of a less obvi-ous left-hemisphere specialization: specialization for thecontrol of ipsilateral movement. Then, the discussionshifts to the three best-documented domains of right-hemisphere specialization: spatial ability, emotion, andmusical ability. Finally, it focuses on two lines of researchon the lateralization of function of cognitive abilities thatare contributing to a new way of thinking about cerebrallateralization of function.

Superiority of the Left Hemisphere in ControllingIpsilateral Movement One unexpected left-hemisphere specialization was revealed by functionalbrain-imaging studies (see Haaland & Harrington, 1996).When complex, cognitively driven movements are madeby one hand, most of the activation is observed in thecontralateral hemisphere, as expected. However, some ac-tivation is also observed in the ipsilateral hemisphere, andthese ipsilateral effects are substantially greater in the lefthemisphere than in the right (Kim et al., 1993). Consistentwith this observation is the finding that left-hemispherelesions are more likely than right-hemisphere lesions toproduce ipsilateral motor problems—for example, left-hemisphere lesions are more likely to reduce the accuracyof left-hand movements than right-hemisphere lesionsare to reduce the accuracy of right-hand movements.

Superiority of the Right Hemisphere in Spatial Ability In a classic early study, Levy (1969) placed athree-dimensional block of a particular shape in either theright hand or the left hand of her split-brain subjects.Then, after they had thoroughly palpated (tactually inves-tigated) it, she asked them to point to the two-dimen-sional test stimulus that best represented what thethree-dimensional block would look like if it were made ofcardboard and unfolded. She found a right-hemispheresuperiority on this task, and she found that the two hemi-spheres seemed to go about the task in different ways. Theperformance of the left hand and right hemisphere wasrapid and silent, whereas the performance of the righthand and left hemisphere was hesitant and often accom-panied by a running verbal commentary that was difficultfor the subjects to inhibit. Levy concluded that the righthemisphere is superior to the left at spatial tasks. This con-clusion has been frequently confirmed (e.g., Funnell, Cor-ballis, & Gazzaniga, 1999; Kaiser et al., 2000), and it isconsistent with the finding that disorders of spatial per-ception (e.g., contralateral neglect—see Chapters 7 and 8)tend to be associated with right-hemisphere damage.

Specialization of the Right Hemisphere for EmotionAccording to the old concept of left-hemisphere domi-nance, the right hemisphere is uninvolved in emotion.This presumption has been proven false. Indeed, analysisof the effects of unilateral brain lesions indicates that theright hemisphere is superior to the left at performingsome tests of emotion—for example, in accurately identi-fying facial expressions of emotion (Bowers et al., 1985).

Superior Musical Ability of the Right HemisphereKimura (1964) compared the performance of 20 right-handers on the standard, digit version of the dichotic lis-tening test with their performance on a version of the testinvolving the dichotic presentation of melodies. In themelody version of the test, Kimura simultaneously playedtwo different melodies—one to each ear—and then askedthe subjects to identify the two they had just heard fromfour that were subsequently played to them through bothears. The right ear (i.e., the left hemisphere) was superiorin the perception of digits, whereas the left ear (i.e., theright hemisphere) was superior in the perception ofmelodies. This is consistent with the observation thatright temporal lobe lesions are more likely to disruptmusic discriminations than are left temporal lobe lesions.

Hemispheric Difference in Memory Early studies ofthe lateralization of cognitive function were premised onthe assumption that particular cognitive abilities reside inone or the other of the two hemispheres. However, the re-sults of research have led to an alternative way of think-ing: The two hemispheres have similar abilities that tendto be expressed in different ways. The study of the lateral-ization of memory was one of the first areas of researchon cerebral lateralization to lead to this modification inthinking. You see, both the left and right hemisphereshave the ability to perform on tests of memory, but theleft hemisphere is better on some tests, whereas the righthemisphere is better on others. Clearly, memory abilitydoes not reside in one of the two hemispheres.

There have been two approaches to resolving the dis-crepant findings of research on the cerebral lateralizationof memory. One approach has been to try to link partic-ular memory processes with particular hemispheres—forexample, it has been argued that the left hemisphere isspecialized for encoding episodic memory (see Chapter11). The other approach (e.g., Wolford, Miller, & Gaz-zaniga, 2004) has been to link the memory processes ofeach hemisphere to specific materials rather than to spe-cific processes. In general, the left hemisphere has beenfound to play the greater role in memory for verbal mate-rial, whereas the right hemisphere has been found to playthe greater role in memory for nonverbal material (e.g.,Kelley et al., 2002). Whichever of these two approachesproves more fruitful—and perhaps both have some ad-vantages—they represent an advance over the tendency tothink that memory is lateralized to one hemisphere.

ISB

N0-558-84818-4



The Left-Hemisphere Interpreter Several lines of ev-idence suggest that the left and right hemispheres ap-proach cognitive tasks in different ways. The cognitiveapproach that is typical of the left hemisphere is attrib-uted to a mechanism that is metaphorically referred to asthe interpreter—a hypothetical neuronal mechanismthat continuously assesses patterns of events and tries tomake sense of them.

The following experiment illustrates the kind of evi-dence that supports the existence of a left-hemisphere in-terpreter. The left or right hemispheres of split-brain pa-tients were tested separately. The task was to guess whichof two lights—top or bottom—would come on next. Thetop light came on 80% of the time in a random sequence,but the subjects were not given this information. Intactcontrol subjects quickly discovered that the top light cameon more often than the bottom one; however, becausethey tried to figure out the nonexistent rule that predictedthe exact sequence, they were correct only 68% of thetime—even though they could have scored 80% if they al-ways selected the top light. The left hemispheres of thesplit-brain subjects performed on this test like intact con-trols: They attempted to find deeper meaning and as a re-sult performed poorly. In contrast, the right hemispheres,like intact rats or pigeons, did not try to interpret theevents and readily learned to maximize their correct re-sponses by always selecting the top light (see Metcalfe,Funnell, & Gazzaniga, 1995; Roser & Gazzaniga, 2004).

What Is Lateralized—Broad Clusters ofAbilities or Individual Cognitive Processes?

I have already alerted you to one way in which examplesof cerebral lateralization are commonly misinterpreted:Contrary to popular thinking, evidence of lateralization

indicates slight hemispheric biases, notall-or-none differences. Now that youhave had the opportunity to digest

some examples of cerebral lateralization of function (seeTable 16.1), I want to issue a second warning—one thathas emerged during the preceding discussion.

You have undoubtedly encountered information aboutcerebral lateralization of function before. Is there a singleeducated person in this society who does not know thatthe left hemisphere is the logical language hemisphereand the right hemisphere is the emotional spatial hemi-sphere? Some of you may even believe that people can beclassified as left-hemisphere or right-hemisphere people.Information like that in Table 16.1 summarizes the resultsof many studies, and thus it serves a useful function if nottaken too literally. The problem is that such informationis almost always taken too literally by those unfamiliarwith the complexities of the relevant research literature.Let me explain.

Early theories of cerebral laterality tended to ascribecomplex clusters of mental abilities to one hemisphere or

the other. The left hemisphere tended to perform betteron language tests, so it was presumed to be dominant forlanguage-related abilities; the right hemisphere tended toperform better on some spatial tests, so it was presumedto be dominant for space-related abilities; and so on. Per-haps this was a reasonable first step, but now the consen-sus among researchers is that this approach is simplistic.

The problem is that categories such as language, emo-tion, musical ability, and spatial ability are each composedof dozens of different individual cognitive activities, andthere is no reason to assume that all those activities asso-ciated with a general English label (e.g., spatial ability)will necessarily be lateralized in the same hemisphere.The inappropriateness of broad categories of cerebral lat-eralization has been confirmed. How is it possible toargue that all language-related abilities are lateralized inthe left hemisphere, when the right hemisphere hasproved superior in perceiving the intonation of speechand the identity of the speaker (Beeman & Chiarello,1998)? Indeed, notable exceptions to all broad categoriesof cerebral lateralization have emerged (see Baas, Aleman,& Kahn, 2004; Josse & Tzourio-Mazoyer, 2004; Ter-vaniemi & Hugdahl, 2003; Vogel, Bowers, & Vogel, 2003).

As a result of the evidence that broad categories ofabilities are not the units of cerebral lateralization, manyresearchers are taking a different approach. They are bas-ing their studies of cerebral lateralization on the work ofcognitive psychologists, who have broken down complexcognitive tasks—such as reading, judging space, and re-membering—into their constituent cognitive processes.Once the laterality of the individual cognitive elementshas been determined, it is possible to predict the lateralityof cognitive tasks based on the specific cognitive elementsthat compose them. The research of Chabris and Kosslyn(1998) is an excellent example of this approach:

Consider the problems of (a) assessing whether one ob-ject is above or below another and (b) assessing whethertwo objects are greater or less than 1 foot apart. Both arespatial tasks, so early theories might have predicted thatthe right hemisphere would be superior at both. Yet bothrequire a verbal response . . ., so perhaps the left hemi-sphere would be better . . . in each case. But if the lefthemisphere is better, could this instead be because of the“analytical” processing required to compare two ele-ments? It is clear that the coarse conceptualizations of-fered by early theories shed little light on even suchapparently similar tasks as these. (p. 8)

Chabris and Kosslyn therefore took a different ap-proach to predicting the laterality of these two simplespatial judgments. They based their research on the cognitive theory of Kosslyn (1994). Kosslyn found evidence that separate processes in the visual systemjudge different types of spatial relations between objects:There is a process for making categorical judgmentsabout spatial relations (e.g., left/right, above/below) andone for making precise judgments of the spatial relations


ISB

N0-

558-

8481

8-4



between objects in terms of their distance and angle from each other. Because the process for makingcategorical spatial judgments is dominant in the left hemi-sphere and the process for making coordinate spatial judg-ments (judgments about distance and angles) is dominantin the right hemisphere, Chabris and Kosslyn predictedthat the left hemisphere would be superior in judgingwhether an object is above or below another and thatthe right hemisphere would be better in judging whethertwo objects are greater or less than a foot apart. Theywere right.

Anatomical Asymmetries of the Brain

Many anatomical differences between the hemisphereshave been documented. Presumably, these result from in-terhemispheric differences in gene expression, some ofwhich have been documented (Sun et al., 2005).

Most of the research effort has focused on trying todocument anatomical asymmetries in areas of cortex thatare important for language. Three of these areas are thefrontal operculum, the planum temporale, and Heschl’sgyrus. The frontal operculum is the area of frontal lobecortex that lies just in front of the face area of the primarymotor cortex; in the left hemisphere, it is the location ofBroca’s area. The planum temporale and Heschl’s gyrusare areas of temporal lobe cortex (see Figure 16.7). Theplanum temporale lies in the posterior region of the lat-eral fissure; it is thought to play a role in the comprehen-sion of language and is often referred to as Wernicke’sarea. Heschl’s gyrus is located in the lateral fissure justanterior to the planum temporale in the temporal lobe; itis the location of primary auditory cortex.

Because the planum temporale, Heschl’s gyrus, andthe frontal operculum are all involved in language-related activities, one might expect that they would allbe unequivocally larger in the left hemisphere than inthe right in most subjects; but they aren’t. The planumtemporale does tend to be larger on the left, but in onlyabout 65% of human brains (Geschwind & Levitsky,1968). In contrast, the cortex of Heschl’s gyrus tends tobe larger on the right, primarily because there are oftentwo Heschl’s gyri in the right hemisphere and only onein the left. The laterality of the frontal operculum is lessclear. The area of the frontal operculum that is visibleon the surface of the brain tends to be larger on theright; but when the cortex buried within sulci of thefrontal operculum is considered, there tends to be agreater volume of frontal operculum cortex on the left(Falzi, Perrone, & Vignolo, 1982).

A word of caution is in order. It is tempting to con-clude that the tendency for the planum temporale to belarger in the left hemisphere predisposes that hemisphereto language dominance. However, thereis little evidence that people with well-developed planum temporale asymme-tries tend to have more lateralized language functions(see Dos Santos Sequeira et al., 2006; Eckert et al., 2006:Jäncke & Steinmetz, 2003). In addition, the fact that sim-ilar structural brain asymmetries have been reported innonhumans suggests that their function is not related tolanguage (see Dehaene-Lambertz, Hertz-Pannier, &Dubois, 2006).

Techniques for imaging the living human brain havemade it easier to look for correlations between particularneuroanatomical asymmetries and specific performancemeasures. Such studies are important because they havethe potential to reveal the functional advantages of cere-bral lateralization. One such study is that of Schlaug andcolleagues (1995). They used structural magnetic reso-nance imaging (MRI) to measure the asymmetry of theplanum temporale and relate it to the presence of perfectpitch (the ability to identify the pitch of individual musi-cal notes). The planum temporale tended to be larger inthe left hemisphere in musicians with perfect pitch thanin nonmusicians or in musicians without perfect pitch(see Figure 16.8).

Most studies of anatomical asymmetries of the brainhave measured differences in gross neuroanatomy, com-paring the sizes of particular gross structures in the leftand right hemispheres. However, anatomists have startedto study differences in cellular-level structure betweencorresponding areas of the two hemispheres that havebeen found to differ in function (see Gazzaniga, 2000;Hutsler & Galuske, 2003). One such study was conductedby Galuske and colleagues (2000). They compared the or-ganization of the neurons in part of Wernicke’s area withits organization in the same part of the right hemisphere.They found that the areas in both hemispheres are organ-


Anterior

Left Right

Posterior

Heschl'sgyrus

Planumtemporale

Heschl'sgyri

Planumtemporale

FIGURE 16.7 Two language areas of the cerebral cortex thatdisplay neuroanatomical asymmetry: the planum temporale(Wernicke’s area) and Heschl’s gyrus (primary auditory cortex).

ISB

N0-558-84818-4



ized into regularly spaced columns of interconnectedneurons and that the columns are interconnected bymedium-range axons. The columns are the same diame-ter in both hemispheres, but they are about 20% furtherapart in the left hemisphere and are interconnected bylonger axons. Presumably, the particular way in which thecolumns are organized in Wernicke’s area is an adaptationfor the processing of language signals.

A similar pattern of findings has emerged from re-search on asymmetries in the hand area of human primary motor cortex (see Hammond, 2002). The handarea in the hemisphere contralateral to the person’s pre-ferred hand tends to be larger and to have more lateralconnections.

Theories of Cerebral Lateralization of Function: Why Did Cerebral Lateralization Evolve?

Several theories have been proposed to explain why cere-bral lateralization of function evolved. All of them arebased on the same general premise: that it is advanta-geous for areas of the brain that perform similar func-tions to be located in the same hemisphere. However,each theory of cerebral asymmetry postulates a differentfundamental distinction between left and right hemi-sphere function. The following are three prominent theo-ries of cerebral asymmetry.

Analytic–Synthetic Theory One theory of cerebralasymmetry is the analytic–synthetic theory. Theanalytic–synthetic theory of cerebral asymmetry holds thatthere are two basic modes of thinking, an analytic mode

and a synthetic mode, whichhave become segregated duringthe course of evolution in the leftand right hemispheres, respec-tively. According to this theory,

. . . the left hemisphere operatesin a more logical, analytical,computerlike fashion, analyzingstimulus information input se-

quentially and abstracting the relevant details, to which itattaches verbal labels; the right hemisphere is primarily asynthesizer, more concerned with the overall stimulusconfiguration, and organizes and processes informationin terms of gestalts, or wholes. (Harris, 1978, p. 463)

Although the analytic–synthetic theory has been thedarling of pop psychology, its vagueness is a problem. Be-cause it is not possible to specify the degree to which anytask requires either analytic or synthetic processing, it hasbeen difficult to subject the analytic–synthetic theory toempirical tests.

Motor Theory A second theory of cerebral asymmetryis the motor theory (see Kimura, 1979). According to themotor theory of cerebral asymmetry, the left hemisphere isspecialized not for the control of speech per se but for thecontrol of fine movements, of which speech is only onecategory. Support for this theory comes from reports thatlesions that produce aphasia also produce other motordeficits (see Serrien, Ivry, & Swinnen, 2006). For example,Kimura and Watson (1989) found that left frontal lesionsproduced deficits in the ability to make both individualspeech sounds and individual facial movements, whereasleft temporal and parietal lesions produced deficits in theability to make sequences of speech sounds and sequencesof facial movements. One shortcoming of the motor the-ory of cerebral asymmetry is that it does not explain whymotor function tends to become lateralized to the lefthemisphere (see Beaton, 2003).

Linguistic Theory A third theory of cerebral asymme-try is the linguistic theory. The linguistic theory of cerebralasymmetry posits that the primary role of the left hemi-sphere is language—in contrast to the analytic–synthetic

Planum temporale

Left Right Left Right

Musician with perfect pitch Nonmusician FIGURE 16.8 The anatomicalasymmetry detected in the planumtemporale of musicians by magneticresonance imaging. In most people,the planum temporale is larger in theleft hemisphere than in the right; thisdifference was found to be greaterin musicians with perfect pitch thanin either musicians without perfectpitch or control subjects. (Adaptedfrom Schlaug et al., 1995.)

ISB

N0-

558-

8481

8-4



and motor theories, which view language as a secondaryspecialization residing in the left hemisphere because ofthat hemisphere’s primary specialization for analyticthought and skilled motor activity, respectively.

The linguistic theory of cerebral asymmetry is based toa large degree on the study of deaf people who use Amer-ican Sign Language (a sign language with a structure sim-ilar to spoken language) and who suffer unilateral braindamage (see Hickok, Bellugi, & Klima, 2001). W.L. is sucha case.

The Case of W.L., the Man Who Experienced Aphasia for Sign Language

W.L. is a congenitally deaf, right-handed male who grewup using American Sign Language. W.L. had a history of cardiovascular disease; and 7 months prior to testing,he was admitted to hospital complaining of right-

side weakness and motor problems. ACT scan revealed a large frontotem-poroparietal stroke. At that time, W.L.’s

wife noticed that he was making many uncharacteristicerrors in signing and was having difficulty understandingthe signs of others.

Fortunately, W.L.’s neuropsychologists managed to ob-tain a 2-hour videotape of an interview with himrecorded 10 months before his stroke, which served as avaluable source of prestroke performance measures. For-mal poststroke neuropsychological testing confirmed thatW.L. had suffered a specific loss in his ability to use andunderstand sign language. The fact that he could produceand understand complex pantomime gestures suggestedthat his sign-language aphasia was not the result of motoror sensory deficits, and the results of cognitive tests sug-gested that it was not the result of general cognitivedeficits (Corina et al., 1992).

The case of W.L. is important because it illustrates adissociation between linguistic (sign) gestures and non-linguistic (pantomime) gestures. The fact that left-hemisphere damage can disrupt the use of sign languagebut not pantomime gestures suggests that the fundamen-tal specialization of the left hemisphere is language.

Evidence of Cerebral Lateralization in Nonhumans

Cerebral lateralization is often assumed to be an exclusivefeature of the hominid brain. One version of the motortheory of cerebral asymmetry is that left-hemisphere spe-cialization for motor control evolved in early hominids inresponse to their use of tools, and then the capacity forvocal language subsequently evolved in the left hemi-sphere because of its greater motor dexterity. This theory

is challenged by reports of handedness in nonhuman pri-mates. For example, Hopkins and colleagues (2005) stud-ied the laterality of four kinds of hand movements inchimpanzees: communicative gestures, reaching, tool use,and coordinated bimanual activity. Overall, there was ev-idence of right-handedness in the sub-jects, but the tendency to use the righthand was significantly greater for thecommunicative gestures, particularly when they were ac-companied by vocalizations. Hopkins and colleagues con-cluded that lateralization for language evolved as aconsequence of a left-hemisphere specialization for com-municative movements, not for tool use. Asymmetries inthe structure of primary motor cortex have been found tobe correlated with handedness in both chimpanzees(Hopkins & Cantalupo, 2004) and monkeys (Phillips &Sherwood, 2005).

Hand preference is not the only evidence of cerebrallateralization of function in nonhuman primates. In somenonhuman primate species, the left hemispheres havebeen found to be dominant for both the production ofcommunicative vocalizations (see Owren, 1990) and theirdiscrimination (Heffner & Heffner, 1984). Moreover, insome nonhuman primate species, the right hemispherehas proven to be superior in the discrimination of facialidentity and expression (Vermeire, Hamilton, & Erd-mann, 1998). All these findings suggest that the evolutionof cerebral laterality preceded the evolution of hominids.

Interestingly, the motor circuit controlling singing inmale canaries is more developed on the left side of theirbrains (see Mooney, 1999). Because birds are not part ofthe evolutionary lineage of humans—birds evolved fromreptiles in a separate lineage—this finding indicates ei-ther that cerebral laterality evolved prior to the split be-tween the two lineages or that the advantages of cerebrallaterality led to its independent evolution in more thanone species.



The chapter is about to switch its focus from cerebral

lateralization to the neural mechanisms of language and

language disorders. This is a good point for you to review

what you have learned by filling in the blanks in the

following sentences. The correct answers are provided

following the last item of the exercise. Be sure to review

material related to your errors and omissions before

proceeding.

1. The cerebral ______ connect the two hemispheres.

2. Left-hemisphere damage plays a special role in both

aphasia and ______.

ISB

N0-558-84818-4


guage has been conducted and inter-preted within the context of this model,reading about the localization of lan-guage without a basic understanding ofthe Wernicke-Geschwind model wouldbe like watching a game of chess withoutknowing the rules.

Historical Antecedents of the Wernicke-Geschwind Model

The history of the localization of language and the his-tory of the lateralization of function began at the samepoint, with Broca’s assertion that a small area in the in-ferior portion of the left prefrontal cortex (Broca’s area)is the center for speech production. Broca hypothesizedthat programs of articulation are stored within this areaand that speech is produced when these programs acti-vate the adjacent area of the precentral gyrus, which con-trols the muscles of the face and oral cavity. Accordingto Broca, damage restricted to Broca’s area should dis-rupt speech production without producing deficits inlanguage comprehension.

The next major event in the study of the cerebral local-ization of language occurred in 1874, when Carl Wernicke(pronounced “VER-ni-key”) concludedon the basis of 10 clinical cases thatthere is a language area in the left tem-poral lobe just posterior to the primary auditory cortex(i.e., in the left planum temporale). This second languagearea, which Wernicke argued was the cortical area of lan-guage comprehension, subsequently became known asWernicke’s area.

Wernicke suggested that selective lesions of Broca’sarea produce a syndrome of aphasia whose symptoms areprimarily expressive—characterized by normal compre-hension of both written and spoken language and byspeech that retains its meaningfulness despite being slow,labored, disjointed, and poorly articulated. This hypo-thetical form of aphasia became known as Broca’s apha-sia. In contrast, Wernicke suggested that selective lesionsof Wernicke’s area produce a syndrome of aphasia whosedeficits are primarily receptive—characterized by poorcomprehension of both written and spoken language andspeech that is meaningless but still retains the superficialstructure, rhythm, and intonation of normal speech. Thishypothetical form of aphasia became known asWernicke’s aphasia, and the normal-sounding but non-sensical speech of Wernicke’s aphasia became known asword salad.

The following are examples of the kinds of speech thatare presumed to be associated with selective damage toBroca’s and Wernicke’s areas (Geschwind, 1979, p. 183):

Broca’s aphasia: A patient who was asked about a dentalappointment replied haltingly and indistinctly: “Yes . . .

41716.4 ■ Cortical Localization of Language: The Wernicke-Geschwind Model

16.4Cortical Localization of Language:The Wernicke-Geschwind Model

So far, this chapter has focused on the functional asym-metry of the brain, with an emphasis on the lateralizationof language-related functions. At this point, it shifts itsfocus from language lateralization to language localiza-tion. In contrast to language lateralization, which refers tothe relative control of language-related functions by theleft and right hemispheres, language localization refers tothe location within the hemispheres of the circuits thatparticipate in language-related activities.

Like most introductions to language localization, thefollowing discussion begins with the Wernicke-Geschwindmodel, the predominant theory of language localization.Because most of the research on the localization of lan-

Scan Your Brainanswers:(1) commissures, (2) apraxia, (3) Broca’s area,

(4) sodium amytal, (5) more, (6) corpus callosum, (7) Sperry, (8) epilepsy,

(9) cross-cuing,(10) left, (11) emotions,(12) left, (13) constituent,

(14) linguistic.

3. Cortex of the left inferior prefrontal lobe became

known as ______.

4. One common test of language lateralization is invasive;

it involves injecting ______ into the carotid artery.

5. Some evidence suggests that the brains of males are

______ lateralized than the brains of females.

6. The ______ is the largest cerebral commissure.

7. ______ received a Nobel Prize for his research on split-

brain patients.

8. Commissurotomy is an effective treatment for severe

cases of ______.

9. The two hemispheres of a split-brain patient can

communicate via an external route; such external

communication has been termed ______.

10. Damage to the ______ hemisphere is more likely to

produce ipsilateral motor problems.

11. Traditionally, musical ability, spatial ability, and ______

have been viewed as right-hemisphere specializations.

12. A neural mechanism metaphorically referred to as

the interpreter is assumed to reside in the ______

hemisphere.

13. Because broad categories of abilities do not appear

to be the units of cerebral lateralization, current re-

searchers have turned to studying the laterality of

______ cognitive processes.

14. Three common theories of cerebral lateralization are

the analytic–synthetic theory, the motor theory, and

the ______ theory.

See the Wernicke-Geschwind modelof Language mod-ule for a clear andvivid explanation ofthe model.


ISB

N0-

558-

8481

8-4



Monday . . . Dad and Dick . . . Wednesday nine o’clock . . . 10 o’clock . . . doctors . . . and . . . teeth.”

Wernicke’s aphasia: A patient who was asked to describea picture that showed two boys stealing cookies reportedsmoothly: “Mother is away here working her work to gether better, but when she’s looking the two boys looking inthe other part. She’s working another time.”

Wernicke reasoned that damage to the pathway connecting Broca’s and Wernicke’s areas—the arcuatefasciculus—would produce a third type of aphasia, onehe called conduction aphasia. He contended that com-prehension and spontaneous speech would be largely in-tact in patients with damage to the arcuate fasciculus butthat they would have difficulty repeating words that hadjust been heard.

The left angular gyrus—the area of left temporal andparietal cortex just posterior to Wernicke’s area—is an-other cortical area that has been implicated in language.Its role in language was recognized in 1892 by Dejerineon the basis of the postmortem examination of one spe-cial patient. The patient suffered from alexia (the inabil-ity to read) and agraphia (the inability to write). Whatmade this case special was that the alexia and agraphiawere exceptionally pure: Although the patient could notread or write, he had no difficulty speaking or under-standing speech. Dejerine’s postmortem examination re-vealed damage in the pathways connecting the visualcortex with the left angular gyrus. He concluded that theleft angular gyrus is responsible forcomprehending language-related vi-sual input, which is received directlyfrom the adjacent left visual cortexand indirectly from the right visualcortex via the corpus callosum.

During the era of Broca, Wer-nicke, and Dejerine, many influen-tial scholars (e.g., Freud, Head, andMarie) opposed their attempts to lo-calize various language-related abil-ities to specific neocortical areas. Infact, advocates of the holistic ap-proach to brain function graduallygained the upper hand, and interestin the cerebral localization of lan-guage waned. However, in the mid-1960s, Norman Geschwind (1970)revived the old localizationist ideasof Broca, Wernicke, and Dejerine,added some new data and insight-ful interpretation, and melded themix into a powerful theory: the Wer-nicke-Geschwind model.

The Wernicke-Geschwind Model

The following are the seven components of theWernicke-Geschwind model: primary visual cortex, an-gular gyrus, primary auditory cortex, Wernicke’s area, ar-cuate fasciculus, Broca’s area, and primary motorcortex—all of which are in the left hemisphere. They areshown in Figure 16.9.

The following two examples illustrate how the Wernicke-Geschwind model is presumed to work (seeFigure 16.10). First, when you are having a conversation,the auditory signals triggered by the speech of the otherperson are received by your primary auditory cortex andconducted to Wernicke’s area, where they are compre-hended. If a response is in order, Wernicke’s area generatesthe neural representation of the thought underlying thereply, and it is transmitted to Broca’s area via the left arcu-ate fasciculus. In Broca’s area, this signal activates the ap-propriate program of articulation that drives the appro-priate neurons of your primary motor cortex andultimately your muscles of articulation. Second, when youare reading aloud, the signal received by your primary vi-sual cortex is transmitted to your left angular gyrus, whichtranslates the visual form of the word into its auditorycode and transmits it to Wernicke’s area for comprehen-sion. Wernicke’s area then triggers the appropriate re-sponses in your arcuate fasciculus, Broca’s area, and motorcortex, respectively, to elicit the appropriate speechsounds.

Broca’sarea

Primarymotorcortex

Primaryauditorycortex

Primaryvisualcortex

Wernicke’sarea

Angulargyrus

Arcuatefasciculus

FIGURE 16.9 The seven componentsof the Wernicke-Geschwind model.

ISB

N0-558-84818-4


16.5Evaluation of the Wernicke-Geschwind Model

Unless you are reading this text from back to front,you should have read the preceding description of theWernicke-Geschwind model with some degree of skepti-cism. By this point in the text, you will almost certainly

41916.5 ■ Evaluation of the Wernicke-Geschwind Model

How the Wernicke-Geshwind model works

Questionheard

Wordsread

Respondingto a heard question

Readingaloud

FIGURE 16.10 How the Wernicke-Geschwind model works in a person who is responding to aheard question and reading aloud. The hypothetical circuit that allows the person to respond toheard questions is in green; the hypothetical circuit that allows the person to read aloud is in black.

Before proceeding to the following evaluation of the

Wernicke-Geschwind model, scan your brain to confirm

that you understand its fundamentals. The correct answers

are provided following the last item of the exercise. Review

material related to your errors and omissions before pro-

ceeding.

According to the Wernicke-Geschwind model, the fol-

lowing seven areas of the left cerebral cortex play a role in

language-related activities:

1. The ______ gyrus translates the visual form of a read

word into an auditory code.

2. The ______ cortex controls the muscles of articulation.

3. The ______ cortex perceives the written word.

Scan Your Brainanswers:(1) angular, (2) primary motor, (3) primary visual,

(4) Wernicke’s, (5) primary auditory, (6) Broca’s, (7) arcuate fasciculus.

4. ______ area is the center for language comprehension.

5. The ______ cortex perceives the spoken word.

6. ______ area contains the programs of articulation.

7. The left______ carries signals from Wernicke’s area to

Broca’s area.

ISB

N0-

558-

8481

8-4



recognize that any model of a complex cognitive processthat involves a few localized neocortical centers joined in a

serial fashion by a few arrows is sure tohave major shortcomings, and you willappreciate that the neocortex is not di-

vided into neat compartments whose cognitive functionsconform to vague concepts such as language comprehen-sion, speech motor programs, and conversion of writtenlanguage to auditory language (see Thompson-Schill,Bedny, & Goldberg, 2005). Initial skepticism aside, the ul-timate test of a theory’s validity is the degree to which itspredictions are consistent with the empirical evidence.

Before we examine this evidence, I want to empha-size one point. The Wernicke-Geschwind model was ini-tially based on case studies of aphasic patients withstrokes, tumors, and penetrating brain injuries. Damagein such cases is often diffuse, and it inevitably encroacheson subcortical nerve fibers that connect the lesion siteto other areas of the brain (see Bogen & Bogen, 1976).For example, illustrated in Figure 16.11 is the extent ofthe cortical damage in one of Broca’s two original cases(see Mohr, 1976).

Effects of Damage to Various Areas ofCortex on Language-Related Abilities

In view of the fact that the Wernicke-Geschwind modelgrew out of the study of patients with cortical damage, itis appropriate to begin evaluating it by assessing its abil-ity to predict the language-related deficits produced bydamage to various parts of the cortex.

Surgical Removal of Cortical TissueThe study of patients in whom discreteareas of cortex have been surgically re-moved has proved particularly informa-tive with regard to understanding thecortical localization of language. This isbecause the location and extent of thesepatients’ lesions can be derived with rea-sonable accuracy from the surgeon’s re-port. The study of neurosurgical patientshas not confirmed the predictions of theWernicke-Geschwind model by anystretch of the imagination. See the sixcases summarized in Figure 16.12.

Surgery that destroys all of Broca’s area but little sur-rounding tissue typically has no lasting effects on speech(Penfield & Roberts, 1959; Rasmussen & Milner, 1975;Zangwill, 1975). Some speech problemswere observed after the removal ofBroca’s area, but their temporal coursesuggested that they were products of postsurgical edema(swelling) in the surrounding neural tissue rather than ofthe excision (cutting out) of Broca’s area per se. Prior tothe use of effective anti-inflammatory drugs, patientswith excisions of Broca’s area often regained conscious-ness with their language abilities fully intact only to haveserious language-related problems develop over the nextfew hours and then subside in the following weeks. Simi-larly, permanent speech difficulties were not produced bydiscrete surgical lesions to the arcuate fasciculus, and per-manent alexia and agraphia were not produced by surgi-cal lesions restricted to the cortex of the angular gyrus(Rasmussen & Milner, 1975).

The consequences of surgical removal of Wernicke’sarea are less well documented; surgeons have been hesitant to remove it in light of Wernicke’s dire predic-tions. Nevertheless, in some cases, a good portion ofWernicke’s area has been removed without lasting lan-guage-related deficits (e.g., Ojemann, 1979; Penfield &Roberts, 1959).

Supporters of the Wernicke-Geschwind model arguethat, despite the precision of surgical excision, negativeevidence obtained from the study of the effects of brainsurgery should be discounted. They argue that the brainpathology that warranted the surgery may have reorgan-ized the control of language by the brain.

Broca’sarea

Area of damage observedin one of Broca’s subjects



FIGURE 16.11 The extent of brain damagein one of Broca’s two original patients. Likethis patient, most aphasic patients have dif-fuse brain damage. It is thus difficult to deter-mine from studying them the precise locationof particular cortical language areas. (Adaptedfrom Mohr, 1976.)

ISB

N0-558-84818-4



Case J.M. No speech difficultiesfor 2 days after his surgery, butby Day 3 he was almost totallyaphasic; 18 days after hisoperation he had no difficulty in spontaneous speech, naming, or reading, but his spelling and writing were poor.

Case H.N. After his operation,he had a slight difficulty in spontaneous speech, but 4 dayslater he was unable to speak; 23 days after surgery, there were minor deficits in spontaneousspeech, naming, and readingaloud, and a marked difficulty in oral calculation.

Case J.C. There were no immediate speech problems;18 hours after his operation he became completely aphasic, but21 days after surgery, only mild aphasia remained.

Case P.R. He had no immediatespeech difficulties; 2 days after hisoperation, he had some language-related problems, but they clearedup.

Case D.H. This operation wasdone in two stages; following completion of the second stage, no speech-related problems were reported.

Case A.D. He had no language-related problems after hisoperation, except for a slightdeficit in silent reading andwriting.

FIGURE 16.12 The lack of permanent disruption of language-related abilities after surgical exci-sion of the classic Wernicke-Geschwind language areas. (Adapted from Penfield & Roberts, 1959.)

Accidental or Disease-Related Brain DamageHécaen and Angelergues (1964) rated the articulation,fluency, comprehension, naming ability, ability to repeatspoken sentences, reading, and writing of 214 right-handed patients with small, medium, or large accidentalor disease-related lesions to the left hemisphere. The ex-tent and location of the damage in each case were esti-mated by either postmortem histological examination orvisual inspection during subsequent surgery. Figure 16.13on page 422 summarizes the deficits found by Hécaenand Angelergues in patients with relatively localized dam-age to one of five different regions of left cerebral cortex.

Hécaen and Angelergues found thatsmall lesions to Broca’s area seldomproduced lasting language deficits and

that those restricted to Wernicke’s area sometimes did notproduce such deficits. Medium-sized lesions did producesome deficits; but in contrast to the predictions of theWernicke-Geschwind model, problems of articulationwere just as likely to occur following medium-sized pari-etal or temporal lesions as they were following compara-ble lesions in the vicinity of Broca’s area. All othersymptoms that were produced by medium-sized lesionswere more likely to appear following parietal or temporallesions than following frontal damage.

Consistent with the Wernicke-Geschwind model, largelesions (those involving three lobes) in the anterior areasof the brain were more likely to be associated with artic-ulation problems than were large lesions in the posteriorareas of the brain. It is noteworthy that none of the 214


ISB

N0-

558-

8481

8-4



subjects displayed syndromes of aphasia that were either to-tally expressive (Broca’s aphasia)or totally receptive (Wernicke’saphasia).

CT and Structural MRI Scansof Aphasic Patients Since thedevelopment of computed to-mography (CT) and structuralmagnetic resonance imaging(MRI), it has been possible to visualize the brain damage ofliving aphasic patients (seeDamasio, 1989). In early CTstudies by Mazzocchi and Vig-nolo (1979) and Naeser and col-leagues (1981), none of theaphasic patients had corticaldamage restricted to Broca’s andWernicke’s areas, and all had ex-tensive damage to subcortical

white matter. Generally, consistent withthe Wernicke-Geschwind model, largeanterior lesions of the left hemisphere

were more likely to produce deficits in language expres-sion than were large posterior lesions, and large posteriorlesions were more likely to produce deficits in languagecomprehension than were large anterior lesions. Also, inboth studies, global aphasia—a severe disruption of alllanguage-related abilities—was associated with very largeleft-hemisphere lesions that involved both anterior andposterior cortex as well as substantial portions of subcor-tical white matter.

The findings of Damasio’s (1989) structural MRI studywere similar to those of the aforementioned CT studies,with one important addition. Damasio found a few apha-sic patients whose damage was restricted to the medialfrontal lobes (to the supplementary motor area and the anterior cingulate cortex), an area not included inthe Wernicke-Geschwind model. Similarly, several CTand MRI studies have found cases of aphasia resultingfrom damage to subcortical structures (see Alexander,1989)—for example, to the left subcortical white matter,the left basal ganglia, or the left thalamus (e.g., Naeseret al., 1982).

Electrical Stimulation of the Cortex andLocalization of Language

The first large-scale electrical brain-stimulation studies ofhumans were conducted by Wilder Penfield and his col-leagues in the 1940s at the Montreal Neurological Insti-tute (see Feindel, 1986). One purpose of the studies wasto map the language areas of each patient’s brain so thattissue involved in language could be avoided during thesurgery. The mapping was done by assessing the re-sponses of conscious patients who were under local anes-thetic to stimulation applied to various points on thecortical surface. The description of the effects of eachstimulation were dictated to a stenographer—this was be-fore the days of tape recorders—and then a tiny num-bered card was dropped on the stimulation site forsubsequent photography.

Figure 16.14 illustrates the responses to stimulation ofa 37-year-old right-handed epileptic patient. He hadstarted to have seizures about 3 monthsafter receiving a blow to the head; at thetime of his operation, in 1948, he hadbeen suffering from seizures for 6 years, despite efforts tocontrol them with medication. In considering his re-

Articulatory disturbances

Difficulties in fluency of speech

Disturbances of verbal comprehension

Disturbances of naming

Disturbances of repetition

Disturbances in reading

Disturbances of writing


FIGURE 16.13 The relative effectson language-related abilities of dam-age to one of five general areas ofleft-hemisphere cortex. (Adaptedfrom Hécaen & Angelergues, 1964.)


ISB

N0-558-84818-4



sponses, remember that the cortex just posterior to thecentral fissure is primary somatosensory cortex and thatthe cortex just anterior to the central fissure is primarymotor cortex.

Because electrical stimulation of the cortex is muchmore localized than a brain lesion, it has been a usefulmethod for testing predictions of the Wernicke-Geschwind model. Penfield and Roberts (1959) publishedthe first large-scale study of the effects of cortical stimu-lation on speech. They found that sites at which stimu-lation blocked or disrupted speech in conscious neuro-surgical patients were scattered throughout a large ex-panse of frontal, temporal, and parietal cortex, ratherthan being restricted to the Wernicke-Geschwind lan-

guage areas (see Figure 16.15 on page 424). They alsofound no tendency for particular kinds of speech distur-bances to be elicited from particular areas of the cortex:Sites at which stimulation produced disturbances of pro-nunciation, confusion of counting, inability to name ob-jects, or misnaming of objects were pretty muchintermingled. Right-hemisphere stimulation almostnever disrupted speech.

Ojemann and his colleagues (see Ojemann, 1983) as-sessed naming, reading of simple sentences, short-termverbal memory, ability to mimic movements of the faceand mouth, and ability to recognize phonemes duringcortical stimulation. A phoneme is the smallest unit ofsound that distinguishes among various words in a

The patient had initial difficulty, but eventually he named a picture of a butterfly.

25

The patient said, “Oh, I know what it is” in response to a picture of a foot. “That is what you put in your shoes.” After termination of the stimulation, he said, “foot.”

26

The patient became unable to name the pictures as soon as the electrode was placed here. The EEG revealed seizure activity in the temporal lobe. When the seizure discharges stopped, the patient spoke at once. “Now I can talk,” he said,and he correctly identified the picture of a butterfly.

28

Tingling in the right thumb and a slight movement

1

Quivering of the jaw in a sidewise manner12

Pulling of jaw to right13

Sensation in the jaw and lower lip14

Tingling in the right side of tongue16

Sensation in right upper lip17

Stimulation, applied while the patient was talking, stopped his speech. After cessation of stimulation, he said that he had been unable to speak despite trying.

23

The patient tried to talk, his mouth moved, but he made no sound.

24

28

1

12 1713 16

24 14

232526

Centralfissure

Lateralfissure

Cut edgeof skull

Portion oftemporal lobethat wasexcised

FIGURE 16.14 The responses of the left hemisphere of a 37-year-old epileptic to electrical stimulation. Numbered cards were placed on the brain during surgery to mark the sites wherebrain stimulation had been applied. (Adapted from Penfield & Roberts, 1959.)

ISB

N0-

558-

8481

8-4


Broca’s aphasia and Wernicke’s aphasia as diagnostic cate-gories, but with an understanding that the syndromes aremuch less selective and the precipitating damage muchmore diffuse and variable than implied by the model(Alexander, 1997). Because of the lack of empirical sup-port for its major predictions, the Wernicke-Geschwindmodel has been largely abandoned by researchers, but it isstill prominent in the classroom and clinic. The last twosections of this chapter focus on an alternative to the Wernicke-Geschwind perspective on the neural mecha-nisms of language: the cognitive neuroscience approach.

16.6Cognitive Neuroscience Approachto Language

The cognitive neuroscience approach is currently domi-nating research on language and its disorders. What is thisapproach, and how does it differ from the traditional per-spective? The following are three related ideas that definethe cognitive neuroscience approach to language. Al-though these ideas were originally premises, or assump-tions, that directed cognitive neuroscience research onlanguage, each one has been supported by a substantialamount of evidence (see Patterson & Ralph, 1999; Saf-fran, 1997).

● Premise 1: Language-related behaviors are mediatedby activity in those particular areas of the brain thatare involved in the specific cognitive processes re-quired for the behaviors. The Wernicke-Geschwindmodel theorized that particular areas of the brain in-volved in language were each dedicated to a specific,but complex, activity such as speech, comprehension,or reading. But cognitive neuroscience research hasfound that each of these activities can itself be brokendown into constituent cognitive processes, which maybe organized in different parts of the brain (Neville &Bavelier, 1998). Accordingly, these constituent cogni-

language; the pronunciation of eachphoneme varies slightly, depending onthe sounds next to it. They found (1) thatthe areas of cortex at which stimulationcould disrupt language extended far be-yond the boundaries of the Wernicke-Geschwind language areas, (2) that eachof the language tests was disrupted bystimulation at widely scattered sites, and(3) that there were major differencesamong the subjects in the organization of language abili-ties (see McDermott, Watson, & Ojemann, 2005).

Because the disruptive effects of stimulation at a particular site were frequently quite specific (i.e., disrupt-ing only a single test), Ojemann suggested that the language cortex is organized like a mosaic, with the discrete columns of tissue that perform a particular func-tion widely distributed throughout the language areas of cortex.

Current Status of the Wernicke-Geschwind Model

Empirical evidence has supported the Wernicke-Geschwind model in two general respects. First, the evi-dence has confirmed that Broca’s and Wernicke’s areasplay important roles in language; many aphasics have dif-fuse cortical damage that involves one or both of theseareas. Second, there is a tendency for aphasias associatedwith anterior damage to involve deficits that are more ex-pressive and those associated with posterior damage toinvolve deficits that are more receptive.

However, the evidence has not been supportive of thespecific predictions of the Wernicke-Geschwind model:

● Damage restricted to the boundaries of the Wernicke-Geschwind cortical areas often has little lasting effecton the use of language.

● Brain damage that does not include any of the Wernicke-Geschwind areas can produce aphasia.

● Broca’s and Wernicke’s aphasias rarely exist in thepure forms implied by the Wernicke-Geschwindmodel; aphasia virtually always involves both expres-sive and receptive symptoms (see Benson, 1985).

● There are major differences in the localization of cor-tical language areas in different individuals (e.g.,Casey, 2002; Schlaggar et al., 2002).

Despite these problems, the Wernicke-Geschwindmodel has been an extremely important theory. It guidedthe study and clinical diagnosis of aphasia for more thanfour decades. Indeed, clinical neuropsychologists still use


Sites at whichstimulation produced a complete arrestof speech

Sites at whichstimulationdisrupted speechbut did not blockit completely

FIGURE 16.15 The wide distribution of lefthemisphere sites where cortical stimulation ei-ther blocked speech or disrupted it. (Adaptedfrom Penfield & Roberts, 1959.)

ISB

N0-558-84818-4


42516.6 ■ Cognitive Neuroscience Approach to language

tive processes, not the general Wernicke-Geschwindactivities, appear to be the appropriate level at whichto conduct analysis. Cognitive neuroscientists typi-cally divide the cognitive processes involved in lan-guage into three categories of activity: phonologicalanalysis (analysis of the sound of language),grammatical analysis (analysis of the structure oflanguage), and semantic analysis (analysis of themeaning of language).

● Premise 2: The areas of the brain involved in languageare not dedicated solely to that purpose (Nobre &Plunkett, 1997). In the Wernicke-Geschwind model,large areas of left cerebral cortex were thought to bededicated solely to language, whereas the cognitiveneuroscience approach assumes that many of the con-stituent cognitive processes involved in language alsoplay roles in other behaviors (see Bischoff-Grethe etal., 2000). For example, some of the areas of the brainthat participate in short-term memory and visual pat-tern recognition are clearly involved in reading as well.

● Premise 3: Because many of the areas of the brain thatperform specific language functions are also parts ofother functional systems, these areas are likely to besmall, widely distributed, and specialized (Neville &Bavelier, 1998). In contrast, the language areas of theWernicke-Geschwind model are assumed to be large,circumscribed, and homogeneous.

In addition to these three premises, the cognitive neu-roscience approach to language is distinguished from thetraditional approach by its methodology. The Wernicke-Geschwind model rested heavily on the analysis of brain-damaged patients, whereas researchers using the cognitiveneuroscience approach also have at their disposal an in-creasing array of techniques—most notably, functionalbrain imaging—for studying the localization of languagein healthy subjects.

Functional Brain Imaging and theLocalization of Language

Functional brain-imaging techniques have revolutionizedthe study of the localization of language. In the lastdecade, there have been numerous PET and fMRI studiesof subjects engaging in various language-related activities(see Martin, 2005; Nakamura et al., 2005). I have selectedtwo to describe to you. As you are about to learn, I se-lected them because they are of high quality, have inter-esting findings, and feature two different approaches. Thefirst is the fMRI study of silent reading by Bavelier andcolleagues (1997); the second is the PET study of objectnaming by Damasio and colleagues (1996).

Bavelier’s fMRI Study of Reading Bavelier and col-leagues used fMRI to measure the brain activity ofhealthy subjects while they read silently. The researchers’

general purpose was not to break down reading into itsconstituent cognitive processes or elements but to get asense of the extent of cortical involvement in reading.

The methodology of Bavelier and colleagues was note-worthy in two respects. First, they used a particularly sen-sitive fMRI machine that allowed them to identify areas ofactivity with more accuracy than in most previous stud-ies and without having to average the scores of severalsubjects (see Chapter 5). Second, they recorded activityduring the reading of sentences—rather than during thesimpler, controllable, and unnatural activities most oftenused in functional brain-imaging studies of language(e.g., listening to individual words).

The subjects in Bavelier and colleagues’ study viewedsentences displayed on a screen. Interposed between peri-ods of silent reading were control periods, during whichthe subjects were presented with strings of consonants.The differences in activity during the reading and controlperiods served as the basis for calculating the areas of cor-tical activity associated with reading. Because of the com-puting power required for the detailed analyses, only thelateral cortical surfaces were monitored.

Let’s begin by considering the findings observed in in-dividual subjects on individual trials, before any averag-ing took place. Three important points emerged from thisanalysis. First, the areas of activity were patchy; that is,they were tiny areas of activity separated by areas of inac-tivity. Second, the patches of activity were variable; that is,the areas of activity differed from subject to subject andeven from trial to trial in the same subject. Third, al-though some activity was observed in the classic Wernicke-Geschwind areas, it was widespread over thelateral surfaces of the brain. The widespread, spotty activ-ity over the left cortex is consistent with the basic prem-ises of the cognitive neuroscience approach and withprevious research—in particular, with brain stimulationstudies of language.

Figure 16.16 on page 426 illustrates the reading-relatedincreases of activity averaged over all the trials and sub-jects in the study by Bavelier and colleagues—as they aretypically reported. The averaging creates the false impres-sion that large, homogeneous expanses of tissue were ac-tive during reading, whereas the patches of activityinduced on any given trial comprised only between 5%and 10% of the illustrated areas. Still, two points are clear:First, although there was significant activity in the righthemisphere, there was far more activity in the left hemi-sphere; second, the activity extended far beyond thoseareas predicted by the Wernicke-Geschwind model to beinvolved in silent reading (e.g., activity in Broca’s area andmotor cortex would not have been predicted).

Damasio’s PET Study of Naming The objective of thestudy of Damasio and colleagues (1996) was to look selec-tively at the temporal-lobe activity involved in namingobjects within particular categories. They recorded PET

ISB

N0-

558-

8481

8-4



Left Hemisphere Right Hemisphere

Active Very active

activity in the left temporal lobes of healthy subjects whilethe subjects named images presented on a screen. The im-ages were of three different types: famous faces, animals,and tools. To get a specific measure of the temporal-lobeactivity involved in naming, the researchers subtractedfrom the activity recorded during this task the activityrecorded while the subjects judged the orientation of theimages. The researchers limited their PET images to theleft temporal lobes of the subjects to permit a more fine-grained PET analysis.

Naming objects activated areas of the left temporallobe outside the classic Wernicke’s language area. Re-markably, the precise area that was activated by the nam-ing depended on the category: Famous faces, animals, andtools each activated a slightly different area. In general,the areas for naming famous faces, animals, and tools arearrayed from anterior to posterior along the middle por-tions of the left temporal lobe.

Damasio and colleagues have not been the only onesto report category-specific encoding of words in the lefttemporal lobes (see Binder et al., 2005; Brambati et al.,2006; Kable, Lease-Spellmeyer, & Chatterjee, 2002).Moreover, the existence of category-specific lexical areasin the left temporal lobes has been supported by theanalysis of aphasic patients with damage in that area.Some patients have naming difficulties that are specificto particular categories (see Kurbat & Farah, 1998), andspecific deficits in naming famous faces, animals, andtools have been shown to correspond to the three areasof the left temporal lobe that were identified by the PETstudy to correspond with these categories (Damasio etal., 1996).

One last point about the cognitive neuroscience ap-proach to language: There has been so much enthusiasmfor the use of functional brain-imagingtechnology to study language that therehas been a tendency among researchersto ignore knowledge gained from the study of brain le-sions. But this is not how science works best: Scienceworks best when new data are added to past data ratherthan supplanting them. For example, significant right-hemisphere activity is virtually always recorded by func-tional brain imaging during language-related activities,suggesting that the right hemisphere plays a significantrole in language; yet lesions of the right hemisphere rarelydisrupt the same activities, suggesting that its role is notcritical. Clearly, a consideration of both types of researchis needed to solve this puzzle (see Price et al., 1999).

16.7Cognitive Neuroscience Approachto Dyslexia

This, the final section of the chapter, looks further at thecognitive neuroscience approach introduced in the pre-ceding section. It focuses on dyslexia, one of the majorsubjects of cognitive neuroscience research.

Dyslexia is a pathological difficulty in reading, onethat does not result from general visual, motor, or intel-lectual deficits. There are two fundamentally differenttypes of dyslexias: developmental dyslexias, dyslexiasthat become apparent when a child is learning to read;


FIGURE 16.16 The areas in which reading-associated increases in activity were observed in thefMRI study of Bavelier and colleagues (1997). These maps were derived by averaging the scores ofall subjects, each of whom displayed patchy increases of activity in 5–10% of the indicated areason any particular trial.

ISB

N0-558-84818-4


Ramus (2004) has proposed a theory that is compati-ble with most of what is known about the developmentaldyslexias and about associated errors in neurodevelop-ment. Ramus agrees with the view that deficits in phono-logical processing are the defining features of acquireddyslexia. The strength of his theory is that it explains whyphonological deficits often occur in conjunction with avariety of subtle sensorimotor deficits that can exacerbatethe condition. Ramus argues that the first stage in the de-velopment of dyslexia is the occurrence of developmentalerrors in auditory areas around the lateral fissure—theseerrors in neural development have been well docu-mented, and one of the genes associated with dyslexiacontrols neuronal migration (see Gal-aburda et al., 2006). In some individuals,the cortical abnormalities extend back tothe thalamus and affect areas there (e.g., the magnocellu-lar layers of the lateral geniculate nuclei). Although thismodel is consistent with most of the available evidence, itdoes not explain why dyslexia is often associated withcerebellar damage (e.g., Justus, 2004). Future researchneeds to identify the precise relations between specificgenes, locations of neurodevelopmental errors, and theresulting cognitive and sensorimotor deficits.

Developmental Dyslexia: Cultural Diversityand Biological Unity

Although it is established that developmental dyslexia isinfluenced by genetic factors and is associated with ab-normalities of brain function, it has long been consideredby many to be a psychological rather than a neural disor-der. Why? Because for many years, those whose thinkingwas influenced by the physiology-or-psychology di-chotomy (see Chapter 2) assumed that developmentaldyslexia could not possibly be a braindisorder because it is influenced by cul-ture. Paulesu and colleagues (2001) re-cently used the cognitive neuroscience approach to drivethe final nail into the coffin of this misguided way ofthinking about dyslexia.

The work of Paulesu and colleagues is based on theremarkable finding that about twice as many Englishspeakers as Italian speakers are diagnosed as dyslexic.This fact has to do with the complexity of the respectivelanguages. English consists of 40 phonemes, which canbe spelled, by one count, in 1120 different ways. In con-trast, Italian is composed of 25 phonemes, which can bespelled in 33 different ways. As a result, Italian-speakingchildren learn to read much more quickly than English-speaking children and are less likely to develop readingdisorders.

Paulesu and colleagues (2000) began by comparingPET activity in the brains of normal English-speakingand Italian-speaking adults. These researchers hypothe-sized that since the cognitive demands of reading aloud

and acquired dyslexias, dyslexias that are caused by braindamage in individuals who were already capable of read-ing. Developmental dyslexia is a widespread problem. Es-timates of the overall incidence of developmental dyslexiaamong English-speaking children range from 5.3% to11.8%, depending on the criteria that are employed to de-fine dyslexia, but the incidence is two to three timeshigher among boys than among girls (Katusic et al.,2001). In contrast, acquired dyslexias are relatively rare.

Developmental Dyslexia: Causes and Neural Mechanisms

Because developmental dyslexia is far more common andits causes are less obvious, most research on dyslexia hasfocused on this form. There is a major genetic compo-nent to developmental dyslexia. The disorder has a her-itability estimate of about 50%, and four genes have sofar been linked to it (see Fisher & Francks, 2006; Gal-aburda et al., 2006).

The problem in identifying the neural mechanisms ofdevelopmental dyslexia is not in discovering pathologicalchanges in the brains of individuals with developmentaldyslexia. The problem is that so many changes have beenidentified that it has been difficult to sort them out (Eck-ert & Leonard, 2003; Roach & Hogben, 2004). As yet, nosingle kind of brain pathology has been found to occur inall cases of developmental dyslexia.

The task of identifying the neural correlates of devel-opmental dyslexia is further complicated by the fact thatthe disorder occurs in various forms, which likely havedifferent neural correlates. Another problem is that read-ing—or its absence—may itself induce major changes in

the brain; thus, it is difficult to deter-mine whether an abnormality in thebrain of a person with developmental

dyslexia is likely to be the cause or the result of the disor-der (see Price & Mechelli, 2005).

Many researchers who study the neural mechanisms ofdyslexia have studied one kind of brain pathology andhave tried to attribute developmental dyslexia to it. Forexample, developmental dyslexia has been attributed toattentional and other sensorimotor deficits caused bydamage to neural circuits linked to the magnocellular lay-ers (see Chapter 6) of the lateral geniculate nuclei (e.g.,Stevens & Neville, 2006). Although many patients withdyslexia do experience a variety of subtle visual, auditory,and motor deficits (Wilmer et al., 2004), many do not (seeRoach & Hogben, 2004). Moreover, even when visual, au-ditory, or motor deficits are present in dyslexic patients,they do not account for all aspects of the disorder. As a re-sult, there is now widespread agreement that dyslexia re-sults from a disturbance of phonological processing (therepresentation and comprehension of speech sounds)—not a disturbance of sensorimotor functioning (see Roach& Hogben, 2004; Shaywitz, Mody, & Shaywitz, 2006).

42716.7 ■ Cognitive Neuroscience Approach to Dyslexia




ISB

N0-

558-

8481

8-4


make often involve the misapplication of common rulesof pronunciation; for example, have, lose, and steak aretypically pronounced as if they rhyme with cave, hose,and beak.

In cases of deep dyslexia (also called phonologicaldyslexia), patients have lost their ability to apply rules ofpronunciation in their reading (i.e., they have lost thephonetic procedure), but they can stillpronounce familiar concrete wordsbased on their specific memories ofthem (i.e., they can still use the lexical procedure). Accord-ingly, they are completely incapable of pronouncing non-words and have difficulty pronouncing uncommonwords and words whose meaning is abstract. In attempt-ing to pronounce words, patients with deep dyslexia try toreact to them by using various lexical strategies, such asresponding to the overall look of the word, the meaningof the word, or the derivation of the word. This leads to acharacteristic pattern of errors. A patient with deepdyslexia might say “quill” for quail (responding to theoverall look of the word), “hen” for chicken (respondingto the meaning of the word), or “wise” for wisdom (re-sponding to the derivation of the word).

I used to have difficulty keeping these two syndromesstraight. Now I remember which is which by remindingmyself that surface dyslexics have difficulty reacting to theoverall shape of the word, which is metaphorically moresuperficial (less deep) than a problem in applying rules ofpronunciation, which is experienced by deep dyslexics.

Where are the lexical and phonetic procedures per-formed in the brain? Much of the research attempting toanswer this question has focused on the study of deepdyslexia. Deep dyslexics most often have extensive dam-age to the left-hemisphere language areas, suggesting thatthe disrupted phonetic procedure is widely distributed inthe frontal and temporal areas of the left hemisphere. Butwhich part of the brain maintains the lexical procedure indeep dyslexics? There have been two theories, both ofwhich have received some support.

One theory is that the surviving lexical abilities ofdeep dyslexics are mediated by activity in surviving partsof the left-hemisphere language areas. Evidence for thistheory comes from the observation of such activity dur-ing reading (Laine et al., 2000; Price et al., 1998). Theother theory is that the surviving lexical abilities of deepdyslexics are mediated by activity in the right hemisphere.The following remarkable case study provides supportfor this theory.

The Case of N.I., the Woman WhoRead with Her Right Hemisphere

Prior to the onset of her illness, N.I. was a healthy girl. Atthe age of 13, she began to experience periods of aphasia,and several weeks later, she suffered a generalized convul-sion. She subsequently had many convulsions, and her

are different for Italian and English speakers, their sub-jects should use different parts of their brains while read-ing. That is exactly what the researchers found. Althoughthe same general areas were active during reading in bothgroups, Italian readers displayed more activity in the leftsuperior temporal lobe, whereas English readers dis-played more activity in the left inferior temporal andfrontal lobes.

Next, Paulesu and colleagues (2001) turned their at-tention to developmental dyslexia. They recorded PETscans of the brains of normal and dyslexic British, French,and Italian university students while the subjects read in-dividual words in their own language. (University stu-dents were used to rule out lack of access to education asa possible confounding factor.) Despite the fact that theItalian dyslexics had less severe reading problems, allthree groups of dyslexics displayed the same pattern ofabnormal PET activity when reading: less than normalreading-related activity in the posterior region of thetemporal lobe, near its boundary with the occipital lobe.Thus, although dyslexia can manifest itself differently inpeople who speak different languages, the underlyingneural pathology appears to be the same.

Cognitive Neuroscience Analysis of ReadingAloud: Deep and Surface Dyslexia

Cognitive psychologists have long recognized that readingaloud can be accomplished in two entirely different ways.One is by a lexical procedure, which is based on specificstored information that has been acquired about writtenwords: The reader simply looks at the word, recognizes it,and says it. The other way reading can be accomplished isby a phonetic procedure: The reader looks at the word,recognizes the letters, sounds them out, and says theword. The lexical procedure dominates in the reading offamiliar words; the phonetic procedure dominates in thereading of unfamiliar words.

This simple cognitive analysis of reading aloud hasproven useful in understanding the symptoms of two re-lated kinds of dyslexia resulting from brain damage (seeCrisp & Ralph, 2006): surface dyslexia and deep dyslexia.Similar syndromes are observed for developmentaldyslexia, but they tend to be less severe.

In cases of surface dyslexia, patients have lost theirability to pronounce words based on their specific mem-

ories of the words (i.e., they have lostthe lexical procedure), but they can stillapply rules of pronunciation in their

reading (i.e., they can still use the phonetic procedure).Accordingly, they retain their ability to pronounce wordswhose pronunciation is consistent with common rules(e.g., fish, river, or glass) and their ability to pronouncenonwords according to common rules of pronunciation(e.g., spleemer or twipple); but they have great difficultypronouncing words that do not follow common rules ofpronunciation (e.g., have, lose, or steak). The errors they




ISB

N0-558-84818-4


Think about It

1. Design an experiment to show that it is possible for asplit-brain student to study for an English exam and ageometry exam at the same time by using a Z lens.

2. The decision to perform commissurotomies on epilepticpatients turned out to be a good one; the decision to per-form prefrontal lobotomies on mental patients (seeChapter 1) turned out to be a bad one. Was this just theluck of the draw? Discuss.

3. Design a fMRI study to identify theareas of the brain involved in compre-hending speech.

4. Why do you think cerebral lateraliza-tion of function evolved?

Themes Revisited

In positioning the themes tabs throughout this chapter, Ilearned something. I learned why this is one of my favoritechapters. This chapter has more themes tabs than any otherchapter—it contributes most to developing the themes ofthe book. Indeed, several passages in this chapter are directlyrelevant to more than one of the themes, which madeplacing the tabs difficult for me.

The clinical implications theme is the most prevalentbecause much of what we know about the lateralization of

function and the localization of languagein the brain comes from the study ofneuropsychological patients.

Because lateralization of function and languagelocalization are often covered by the popular media, they

have become integrated into pop culture,and many widely held ideas about thesesubjects are overly simplistic. In this

chapter, thinking clearly tabs mark aspects of laterality andlanguage that require particularly clear thinking.

Evolutionary analysis has not played amajor role in the study of the localizationof language, largely because humans arethe only species with well-developed language. However,it has played a key role in trying to understand why cere-bral lateralization of function evolved in the first place, andthe major breakthrough in understanding the split-brainphenomenon came from comparative research.

The neuroplasticity theme arose dur-ing the discussion of the effect of earlybrain damage on speech lateralization and during the discussion of the neuro-developmental bases of developmentaldyslexia. Damage to the brain alwaystriggers a series of neuroplastic changes,which can complicate the study of thebehavioral effects of the damage.

them into sounds; she can read concrete familiar words;she cannot pronounce even simple nonsense words (e.g.,neg); and her reading errors indicate that she is readingon the basis of the meaning and appearance of wordsrather than by translating letters into sounds (e.g., whenpresented with the word fruit, she responded, “Juice . . .it’s apples and pears and . . . fruit”). In other words, shesuffers from a severe case of deep dyslexia (Patterson,Vargha-Khadem, & Polkey, 1989).

The case of N.I. completes the circle: The chapterbegan with a discussion of language and lateralization offunction, and the case of N.I. concludes it on the samenote.

speech and motor abilities deteriorated badly. CT scansindicated ischemic brain damage to the left hemisphere.

Two years after the onset of her disorder, N.I. was ex-periencing continual seizures and blindness in her rightvisual field, and there was no meaningful movement orperception in her right limbs. In an attempt to relievethese symptoms, a total left hemispherectomy was per-formed; that is, her left hemisphere was totally removed.Her seizures were totally arrested by this surgery.

The reading performance of N.I. is poor, but she dis-plays a pattern of retained abilities strikingly similar tothose displayed by deep dyslexics or split-brain patientsreading with their right hemispheres. For example, sherecognizes letters but is totally incapable of translating

429Think about It





See Hard Copy foradditional readingsfor Chapter 16.

Studying for anexam? Try thePractice Tests forChapter 16.

ISB

N0-

558-

8481

8-4



Cerebral commissures (p. 401)Lateralization of function

(p. 401)Split-brain patients (p. 401)Commissurotomy (p. 401)

16.1 CerebralLateralization ofFunction: Introduction

Aphasia (p. 402)Broca’s area (p. 402)Apraxia (p. 402)Dominant hemisphere (p. 402)Minor hemisphere (p. 402)Sodium amytal test (p. 402)Dichotic listening test (p. 402)Dextrals (p. 403)Sinestrals (p. 403)

16.2 The Split Brain

Corpus callosum (p. 404)Scotoma (p. 404)Cross-cuing (p. 407)

Helping-hand phenomenon(p. 408)

Visual completion (p. 408)Chimeric figures test (p. 408)Z lens (p. 409)

16.3 Differencesbetween the Left andRight Hemispheres

Interpreter (p. 413)Frontal operculum (p. 414)Planum temporale (p. 414)Heschl’s gyrus (p. 414)

16.4 CorticalLocalization of Language: The Wernicke-Geschwind Model

Wernicke’s area (p. 417)Expressive (p. 417)Broca’s aphasia (p. 417)Receptive (p. 417)Wernicke’s aphasia (p. 417)

Word salad (p. 417)Arcuate fasciculus (p. 418)Conduction aphasia (p. 418)Angular gyrus (p. 418)Alexia (p. 418)Agraphia (p. 418)Wernicke-Geschwind model

(p. 418)

16.5 Evaluation of the Wernicke-Geschwind Model

Global aphasia (p. 422)Phoneme (p. 423)

16.6 CognitiveNeuroscience Approach to Language

Phonological analysis(p. 425)

Grammatical analysis(p. 425)

Semantic analysis (p. 425)

16.7 CognitiveNeuroscience Approach to Dyslexia

Dyslexia (p. 426)Developmental dyslexias

(p. 426)Acquired dyslexias (p. 427)Lexical procedure (p. 428)Phonetic procedure (p. 428)Surface dyslexia (p. 428)Deep dyslexia (p. 428)Hemispherectomy (p. 429)

Key Terms

Need some helpstudying the keyterms for thischapter? Check out the electronicflash cards forChapter 16.

ISB

N0-558-84818-4


Documents

Right Brain of Language - University of Phoenix