Email Calvin || Email Ojemann  || Book's Table of Contents || Calvin Home Page  


William H. Calvin and George A. Ojemann, Inside the Brain:  Mapping the Cortex, Exploring the Neuron (New American Library, 1980), chapter 4. See also

copyright ©1980 by William H. Calvin and George A. Ojemann
Webbed Reprint Collection
This 'tree' is really a pyramidal neuron of cerebral cortex.  The axon exiting at bottom goes long distances, eventually splitting up into 10,000 small branchlets to make synapses with other brain cells.
William H. Calvin

University of Washington
Seattle WA 98195-1800 USA


Subdividing Language Cortex: Naming, Sequencing, Syntax, and Short-term Memory

©1980 by 
William H. Calvin and 
George A. Ojemann


Stroke victims have taught science a great deal: by their own descriptions of their difficulties and from their patient cooperation with neurologists, a rough picture of cortical function has been built up over the years. Epileptic seizures have provided another way of guessing at the function of a particular portion of the cortex, but electrical stimulation of the exposed cortical surface gives more precise localization than strokes or seizures. The exposure of the brain during an epilepsy operation under local anesthesia provides a rare opportunity to explore the organization of the human language cortex, and Neil has agreed to help out.

The "slide show" that was used to define the language portion of the cerebral cortex for the neurosurgeon is now finished. A new tray of slides for the research study is loaded into the projector. A TV camera positioned to watch Neil's face will record the action on videotape for later analysis.

"Neil, we're all set to try those slides where you have to stick out your tongue. All comfortable?" asks the neurosurgeon.

"Sure. I never thought that I'd be on television, sticking out my tongue."

Neil's task, which he rehearsed the night before, is seemingly trivial. Each slide shows drawings of three faces: for example, a face with lips in a kissing position, followed by a face with the tongue sticking out straight, then to the right. Neil is to mimic the three faces in sequence, so he puckers up his lips and then sticks out his tongue straight and then off to the right. The next slide shows further instructions in the form of three more faces. As with the language testing, electrical stimulation is sometimes applied, sometimes not. When the stimulation is applied at certain sites, Neil has trouble producing the sequence of facial movements. A map is thus made of the various sites in Neil's brain where sequential oral-facial movements are disrupted by stimulation.

The movements often involve both left and right sides of the face simultaneously; although only left brain is being stimulated, that is capable of disrupting the sequence of these movements. Stimulation of the right brain, in other patients whose right hemisphere is being operated upon for an epileptic focus, does not do this. Thus, the control of sequential oral-facial movements is also lateralized to the left side, just like language.

Such research was inspired by studies of stroke patients with language difficulties. Many of these stroke victims can mimic a single facial position, such as sticking their tongue out, but they cannot mimic a sequence of them such as Neil can. It was suspected that there was an area in the left brain which is important to the sequencing of motor tasks, probably hand movements as well as face movements. A deaf person who communicates using sign language may I be just as impaired by a left brain stroke as a speaking person.

"Okay, Neil. Now we come to the nonsense words. Tell us the letter which changes."

Neil now listens to a series of nonsense words in which one speech sound (phoneme) has been changed: "adma, akma, abma . . ." He identifies the changed phoneme: "k," "b," and so on. This ability to identify phonemes can be disrupted by stimulating certain sites in the language cortex, so that Neil utters an erroneous consonant.

Neil is also asked to read and complete a series of simple sentences which are shown on slides, such as "If my son is late for class again, the principal . . . " When electrical stimulation is applied at certain sites, Neil will make peculiar errors, such as "If my son will getting late today, he'll see the principal." These errors seem to involve only the grammar of the sentence, not its content.

The inability to name objects, but with the ability to speak still functioning ("This is a . . ., is a . . ., I just can't. . . "), is one way to define the language cortex. The study that Neil has just completed is designed to determine whether sites exist which disrupt phoneme discrimination, or grammar, or sequential facial movements, or memory. If so, are these aspects altered from the same or different sites?

A surprising result is that at sites where phoneme recognition can be disrupted, stimulation will also interfere with the mimicry of sequential oral-facial movements. The converse is also true. This finding supports an idea known as the "motor theory of speech perception," Studies have been done of the way manipulation of acoustical cues alters speech understanding; they indicate that understanding of a speech sound has more in common with the movements with which the sound is spoken than with its acoustical properties. Based on those studies,' it has been proposed that understanding the spoken word depends upon the brain making an internal model for how to speak the word it hears. The study which Neil has just completed identifies an area of the brain where that internal modeling might occur: this region has a role in both speech understanding and speech production.

At other sites in the language cortex, only one of these additional tests is disrupted. Sites disrupting memory rarely show any other effect. Sites showing grammar disruption show no other effects, at least not for any of the aspects of language tested thus far in patients. At the level of resolution provided by surface electrical stimulation, the language cortex seems to put different aspects of language in different places. The exact location of such subdivisions varies from patient to patient (just as the map of the motor strip may vary), but their location relative to each other is generally predictable.

Such findings will likely force a revision of the traditional ideas about brain organization for language which have been inferred from stroke victims; these have emphasized separate regions for receiving language and for producing speech. The new findings suggest that some areas of brain are common to expressive and receptive aspects of language. The major subdivisions of language cortex instead seem to be this common area for sequencing movements and understanding speech, and a surrounding area for memory. Both of these major subdivisions are to be found in frontal as well as parietal-temporal lobes. So it isn't too surprising that careful testing of patients with traditional motor aphasia from frontal damage often shows a subtle defect in understanding as well; that is likely from damage to the cortical subdivision common to sequencing face movements and speech understanding. Indeed, damage to this subdivision may underlie all language disturbance. Additional damage, to the memory subdivision, likely leads to a predominantly expressive aphasia. Additional damage to cortex involved in oral-facial motor control probably leads to a predominantly motor aphasia. Language, then, seems to arise in the interaction between these two major subdivisions, with at least some specialized aspects of language such as grammar located at cortical sites between the sequential-movement/speech-understanding and the memory subdivisions.

In the early nineteenth century, neurologists proposed a rather detailed map for functional localization in the cerebral cortex, even down to having a separate area for grammar. Their views fell into disfavor and they were derided as "diagram makers" by those holding a contrary view: to use grammar as an example, the critics would have said that broad areas of brain worked cooperatively to produce grammar, so that no one area was a specialist in grammar. One of the early leaders of the Viennese localization school, Franz Joseph Gall, went on to propose that various regions could be detected by feeling the bumps on the skull, which did not help to keep the idea of detailed maps respectable. While the nineteenth-century maps are not being resurrected, since they were based largely upon speculation rather than observation, the new information from patients such as Neil show that the old ideas were partially right.

Except for the cooperation and participation of patients such as Neil, it is difficult to imagine how such details of the language cortex would be discovered. Epileptic seizures seldom stay confined to such small regions. Strokes may be that small, but matching up symptoms with a brain site is difficult for small strokes: initially, a lot of surrounding brain is made temporarily nonfunctional by pressure (the killed area swells); thus, the initial symptom may be produced by neighboring regions. Later, other areas may take over the functions of the damaged region so that the symptoms disappear, confounding attempts to identify the region's function.

So taking some extra time in the middle of an epilepsy operation is one of the best ways to get the information. Who decides whether to do it? Finally, the patient. But before asking the patient to participate, the researcher must first take a description of the study to a special committee (often called a review committee on human experimentation) whose members are physicians, nurses, and various types of professors and laypersons. This committee asks about the risks (such as prolonging the operation, which might increase the risk of infection). Do the risks outweigh the potential benefits to the patient (knowing more about the functional localization may help the surgeon to better plan the region of the brain to be removed) and to society (fO T example, potential benefits of new knowledge to dyslexic children or stroke victims)? Can't the study be done in animals instead of humans? (Not language studies.) The investigator must also show the committee the written consent form which the patient will be given to sign. Such committees often ask the investigator to rewrite it, to explain things in simpler language on the written form and to make the risks and potential benefits clearer. Such carefully monitored human experimentation is involved not only in the study of abilities unique to man, but also when new treatments (surgery, drugs) developed in animal research are first applied to patients.

Continue to CHAPTER 5

Notes and References for this chapter  
Book's Table of Contents  

Copyright 1980 by
William H. Calvin and George A. Ojemann

Email Calvin  ||  Email Ojemann  || Calvin Home Page