Modularity and Connectivity

To understand how brains might work when people are performing creative acts, we have to understand the modular organization of the brain, and the modular organization of the brain is at least in part dependent on its anatomy. The cerebral cortex is divided into a left and right hemisphere, and these two hemispheres are connected by a structure called the corpus callosum. The corpus callosum allows information to be transferred from one hemisphere to the other (see Figure 1.1).

Each hemisphere has four major lobes: frontal, temporal, parietal, and occipital (see Figure 1.1). Because most knowledge is stored in the cerebral cortex, it is important to have a large cortex. One way of increasing the relative size of the cortex without increasing the overall size of the brain is to have convolutions, and the human brain is more convoluted than the brain of any other organism. A convoluted brain has mountains called gyri, valleys called sulci, and gorges called fissures. Some of the major gyri of the brain that we will be discussing are diagrammed in Figure 3.5.

Information is carried into the brain from structures such as the eyes, ears, skin, and joints. Before this information reaches the cortex it goes through a relay station in the center of the brain called the thalamus (see Figure 3.8). The visual-sensory information then goes from the optic thalamus to the striate cortex (primary visual area or V1), which is located in the occipital cortex (see Figure 1.1). Joint position and touch information goes from the body to the thalamus and from

Superior Frontal Gyrus

Inferior Fronta Gyrus (Pars Triangularis)

Sylvian Fissi (Horizontal Li

Superior Frontal Gyrus

Inferior Fronta Gyrus (Pars Triangularis)

Sylvian Fissi (Horizontal Li

i—Inferior Temporal Gyrus

—Middle Temporal Gyrus — Superior Temporal Gyrus

Figure 3.5. Diagram of some of the major gyri and fissures on a lateral view of the brain.

'—CaIcarine Cortex

—f \-Precentral Gyrus (Motor Strip)

-h^v-Postcentral Gyrus (Sensory Strip)

r-tr-Supramarginai Gyrus

—V-Angular Gyrus i—Inferior Temporal Gyrus

—Middle Temporal Gyrus — Superior Temporal Gyrus

Figure 3.5. Diagram of some of the major gyri and fissures on a lateral view of the brain.

the thalamus to the postcentral gyrus (primary somatosensory area or S1) (see Figure 1.1) and auditory information goes from the ears to the auditory thalamus and then is relayed to an area on the dorsal (top) surface of the temporal lobe, called Heschl's gyrus (primary auditory area or A1) (see Figure 1.1).

The function of these primary areas is to perform analyses of incoming stimuli. For example, the left visual (striate) cortex is important for detecting stimuli that occur on the right side of space (e.g., stimuli that fall on the nasal portion of the retina in the right eye and the temporal portion of the retina in the left eye). The primary cortex analyzes and detects changes in brightness (edges) that are oriented in specific directions and are positioned in specific portions of space. After the primary sensory cortices perform this analysis of the incoming information, this partially processed information is passed to areas called modality-specific association cortices. For the brain to derive meaning when a person sees an object, it must put these lines or edges together to form a percept of the viewed object's shape. Patients with lesions of the brain who can not recognize objects and who can not draw these objects or even match them to other examples of the same object or pictures of the same object have a disorder called appercep-tive agnosia. Although these patients are not blind, they cannot develop a percept of the object because they cannot put all the edges they see together to form a shape. These patients often have injury to the visual association areas that surround the primary visual areas. These patients' problems are not with naming or activating concepts since when these same objects are presented in other modalities, such as touch, the patients are able to name the object, describe the object, or show how the object is used.

In the clinic we see patients who cannot recognize objects (or faces) but are able to draw these objects or match them to similar objects. This disorder is called associative agnosia. For example, when shown a picture of a hammer, these patients might be unable to name it, describe its use, or demonstrate how it is used, but they could draw the hammer. There are two mechanisms that can produce this form of agnosia. One reason that these patients may fail to recognize objects is that they have a modality-specific memory failure. Visual association areas, primarily in the posterior (back) ventral (bottom) portions of the brain in the temporal and occipital lobes, can store memories of objects that a person has previously seen (see Figure 1.1). These iconic representations of objects, also called structural descriptions, are stored independently from conceptual-semantic representations, which are important to understanding the meaning of the object. For example, if I showed you an abstract drawing, took away this drawing, and then presented it again to you some time later, together with foils (other similar drawings), you should have little difficulty recognizing the drawing that I originally showed you. If I asked you the meaning of this drawing you might say, "It has no meaning." Thus, modality-specific portions of our cerebral cortex do store modality-specific sensory memories, and if these areas are damaged we may be unable to recognize objects or faces.

There are patients who have associative visual agnosia who cannot name or recognize the meaning of objects or their use, but patients with this form of associative agnosia can form percepts (can draw or match, or both, the sample) and also maintain their visual memories or structural descriptions of objects. The means by which we test their visual memories is to show these patients pictures of objects. Some of these pictures are real objects and others are pseudo-objects. Although these patients can discriminate between the real (which they have seen before) and pseudo-objects, they cannot tell the names of these objects or how to use the real objects. Thus these patients have intact memories of objects (structural descriptions), but these memories cannot successfully access the language and conceptual areas of the brain.

The visual, auditory, and tactile association areas all send projections that meet in the posterior inferior parietal lobe and in the posterior portion of the temporal lobes (see Figure 3.6). Thus, these areas are called

Figure 3.6. Diagram demonstrating that the primary auditory area (A1), the primary visual area (V1), and the primary somatosensory area (S1) all project to their modality-specific association areas (A2, S2) and visual association areas in the ventral temporal lobe (VTL) and the superior parietal lobe (SPL). These modality-specific association areas then all project to or converge in the inferior parietal lobe (IPL).

Figure 3.6. Diagram demonstrating that the primary auditory area (A1), the primary visual area (V1), and the primary somatosensory area (S1) all project to their modality-specific association areas (A2, S2) and visual association areas in the ventral temporal lobe (VTL) and the superior parietal lobe (SPL). These modality-specific association areas then all project to or converge in the inferior parietal lobe (IPL).

multimodal or polymodal association areas. Humans' inferior parietal lobe, including the angular gyrus and the supramarginal gyrus (see Figure 3.5), are not present in monkeys' brains. The meeting of all the modality-specific association areas in the posterior parietal and temporal lobe allow humans to perform functions that cannot be performed by monkeys. For example, even in the absence of training, you would have no difficulty pointing to the second array if I, after making three short sounds and then two long sounds and showing you a piece of paper with the following choices, nonverbally indicated that you should point:

(1)__.. _ (2) ...__(3) _._._. Without extensive training, monkeys would have trouble performing this task. The reason people can easily perform this task is that they can make cross-modal (auditory to visual) associations. Cross-modal associations allow us to develop symbols. Letters are symbols of phonemes. A series of phonemes or words are symbols for actions and objects. Numbers are symbols of quantity.

Hemispheric Asymmetries

Several decades ago Marcel Kinsbourne said to me, "It is a good thing humans have two hemispheres because they love dichotomies."

I replied, "Perhaps humans love dichotomies because they have two hemispheres." Some of the dichotomies that have been used to express the differences in functions between the hemispheres include the following: left linguistic and right nonlinguistic, left categorical and right coordinate, left analytic and right gestalt, left sequential and right simultaneous, left rational and right emotional, and left propositional and right appositional. Many discussions of the lateral asymmetries of hemispheric processing suggest that, depending on the type of cognitive process required, either the right or left hemisphere mediates an activity. I suspect, however, that this either-or proposition is incorrect, and that localization of function or modularity permits parallel processing, and parallel processing confers a processing advantage. For example, when PET first was used to study the brain, investigators injected a glucose radioisotope into the blood of normal participants while they were listening to someone speak. When neurons in an area of the brain are active they use more glucose than areas that are inactive, and with an increase of glucose use, the radioactivity emanating from these active areas increases. These investigators found that not only did a portion of the left hemisphere's language area, called Wernicke's area, show an increase (activation), but also a similar area in the right hemisphere also activated. Because many people in neurology still did not believe in localization of function, and they knew that I was a localizationist, they asked me to discuss the results of one of these studies at an annual meeting of the American Academy of Neurology. I mentioned that we recently reported that patients with right temporal-parietal lesions, unlike those with left hemisphere lesions, can understand what is said, or the propositional message, but not how it is said, the emotional tone of voice or affective prosody. In contrast, we demonstrated that people with left temporal-parietal lesions might not be able to understand what is said but can understand how it is being said (emotional prosody). When a normal person listens to another person speak, he or she listens to and processes what is being said not only by the words but also by the tone of voice. Whereas the left hemisphere mediates the comprehension of words, the right hemisphere mediates the understanding of the emotional tone of voice. The ability of each hemisphere to perform a different analysis of the speech (parallel processing) allows normal people to listen simultaneously to both the propositional verbal message and the prosodic emotional message. Thus, when normal participants listen to someone speak, the reason both the left and right hemispheres activate is that these participants are performing simultaneous processing.

Many creative endeavors also probably require parallel processing of the two hemispheres and interactions between the two hemispheres.

Some types of creative acts, however, might depend on the functions or portions of one hemisphere more than the other. Thus, in the next two sections we briefly discuss some of the major functions of the left and right hemispheres. Unfortunately, an exhaustive discussion of all the lateralized functions is beyond the scope of this book, and I suggest that readers who want to learn more details about these lateralized systems read portions of a text edited by Edward Valenstein and me called Clinical Neuropsychology (fourth edition, 2003).

The Left Hemisphere

Studies of patients with injuries to the left hemisphere have revealed that these patients have deficits in propositional language. Those with posterior lesions (e.g., Wernicke's area or the posterior portion of the superior temporal lobe) have problems with verbal comprehension (see Figure 3.3). Patients with injuries to regions that are above, behind, or below Wernicke's area often have problems with finding words. People with anterior lesions (i.e., in the region of Broca's area) become nonfluent and have trouble with verbal expression (see Figure 3.2). Patients with lesions in the region of the angular gyrus in the left parietal lobe have trouble with reading and writing (see Figure 3.5). People with injuries to the left angular gyrus also have trouble with calculations—such as adding, subtracting, multiplying, and dividing—and trouble knowing how to perform learned skilled acts and how to solve mechanical problems. The left hemisphere is also important for focused attention and appears to be superior to the right in categorical reasoning.

The Right Hemisphere

Although lesion, behavioral (dichotic listening and visual half field), functional imaging, and electrophysiological studies provide converging evidence that the left hemisphere of right-handed people appears to be dominant for programming skilled movements, calculating, speaking, and understanding speech as well as related functions (such as reading and writing), the right hemisphere appears to be important in spatial cognition. For example, lesion and behavioral studies using a test in which participants have to compute and recognize the spatial relationship of two lines have revealed that the right hemisphere appears to be dominant (Hamsher, Levin, & Benton, 1979). The right hemisphere is also better at determining if pairs of photographic portraits, which are taken at different angles, of people not known to the participants are the same or different people (Benton, 1990). The right hemisphere also plays a dominant role in recognizing photographs of objects that are taken from unusual angles, such as a picture of a bucket taken from directly above.

Although the right hemisphere appears to be dominant at mediating visuospatial skills, studies of patients' ability to draw or copy complex diagrams has revealed that lesions to both the right and left hemisphere impair drawing. This deficit is called constructional apraxia. Some investigators have suggested that constructional apraxia associated with right-hemisphere injury is different from that associated with left-hemisphere injury. For example, when asked to draw a cube, patients with right-hemisphere injuries appear to have trouble drawing the relationships between the lines or angles and the people with the left-hemisphere lesions appear to have problems with organizing the elements of the cube.

In addition to mediating spatial skills, the right hemisphere appears to be important in mediating several forms of nonpropositional speech. I briefly mentioned previously that the right hemisphere mediates the comprehension of emotion prosody, but it is also important for the expression of emotional prosody (Heilman, Nadeau, & Beversdorf, 2003; Ross, 1981; Tucker, Watson, & Heilman, 1977). The comprehension and expression of facial emotions also appear to be mediated primarily by the right hemisphere. In the clinic we have seen patients who have a loss of fluent propositional speech from left frontal lesions, either in Broca's area or above it, but when some of the these patients are asked to perform automatic speech, such as reciting the Pledge of Allegiance or the Lord's Prayer, they can do so fluently. Some of these nonfluent patients can also fluently sing the words to songs and also curse.

One explanation for this preserved automatic speech is that it is being mediated by the uninjured right hemisphere. Support for this postulate came when I was visiting Israel and saw a patient with Lynn Speedie and Eli Wertman (Speedie, Wertman, Tair, & Heilman, 1993). This man was an Orthodox Jew who awoke to find that he could not chant his morning prayers. This man was now about 80 years old and had been chanting these same prayers since he was a boy. When we examined him and asked him to sing, he also had trouble. He was able to tell this entire story, however, using fluent propositional speech. When he received a CT scan it demonstrated that he had a stroke in the right frontal lobe and right basal ganglia. We also reported a woman who had a progressive loss of the ability to express emotions, both by speech prosody and by emotional facial expressions (Ghacibeh & Heilman, 2003). Her MRI showed focal atrophy of her right frontal lobe. Borod and her coworkers (2000) provided evidence that, when compared with nonemotional words processed by the language-dominant left hemisphere, the right hemisphere also processes emotional words.

Brownell and his coworkers (Brownell, Potter, Bihrle, & Gardner, 1986; Brownell, Simpson, Bihrle, Potter, & Gardner, 1990) performed a series of studies that demonstrated that the right hemisphere mediates metalinguistic functions, such as the comprehension of metaphor. For example, if someone with a right-hemisphere lesion was asked to interpret the sentence "He had a heavy heart" and then given two possible interpretations such as, "He was an athlete and because of all the exercise his heart became muscular" or "He was grieving," they might select the former sentence.

Some studies also suggest that the right hemisphere might play a special role in music. Milner (1962) studied a group of patients who had either their left or their right temporal lobes removed because they had medically uncontrollable seizures, and their physicians were attempting to remove the portion of the brain from which these seizures emanated. She assessed these peoples' musical skills with the Seashore Test of Musical Abilities and found that whereas the people who had a removal of the left hemisphere's temporal lobe appeared to have a deficit of ability to recognize and recall rhythm, the people with removal of their right temporal lobe had more problems with melody. Other investigators, however, have not been able to completely replicate these results (Kester, Saykin, Sperling, & O'Conner, 1991).

Patients with right-hemisphere strokes often demonstrate a disorder where they appear to be severely inattentive or even unaware of stimuli that are presented on the left side of space. Although they are more likely to detect stimuli that are presented on the right side of space than those presented on the left side, these patients are also somewhat inattentive to right-side stimuli. To account for this asymmetry, we suggested and provided evidence (Heilman & Van den Abell, 1980) for the postulate that whereas the left hemisphere can attend to stimuli in the right side of space, it cannot attend to left-side stimuli. In contrast, the right hemisphere can attend both to the opposite (left) side of space and to the ipsilateral (right) side. Thus, when the left hemisphere is injured, these patients' right hemisphere can mediate attention in ipsilateral hemispace. When the right hemisphere is injured, however, the left hemisphere can attend somewhat to contralateral right hemispace, but because it is only this injured right hemisphere that can attend to the left-side space, the left side of space goes unattended or neglected.

Based on the above discussion, it appears that overall the right hemisphere appears to be dominant for mediating attention. When presented with a novel or important stimulus, an organism must be alert. When an organism is alert or aroused, its cerebral cortex is physiologically prepared to process incoming stimuli. The right hemisphere not only is dominant for making attentional computations, such as where in space a person needs to direct their attention, as well as determining the significance of a stimulus, but also appears to control the level of arousal (Heilman, Schwartz, & Watson, 1978).

Some tasks require that a person pay attention to details (focused attention), and other tasks require that a person attend to an overall configuration (global attention). For example, look at Figure 3.7, which is called a Navon figure. Now, attempt to find the letter A and then find the letter H. To find the letter A you used focused attention, but when you attempted to find the letter H you used global attention. Research has revealed that whereas the left hemisphere appears to be important in mediating focal attention, the right hemisphere appears to mediate global attention (Barrett, Beversdorf, Crucian, & Heilman, 1998; Robertson & Lamb, 1991).

When performing creative acts, one must find the thread that unites, or see the unity in what appears to be diversity. If one looks at the Navon figure (see Figure 3.7), it is apparent that the letter H is composed of many smaller letters, but to see the H a person needs to see the thread that unites and thus has to use a global strategy. Hence, at the innovative stage of creativity, the right hemisphere-mediated global attentional systems might be more important that than the left hemisphere-mediated focal systems.

During the verification-production stage of creativity, a scientist or artist also has to concentrate on details and thus must use the left hemisphere-mediated, focused system. As I mentioned previously, there are patients with left-side neglect who are unaware of stimuli that are presented on the side of space that is opposite to their right hemispheric lesion. Denny-Brown and Chambers (1958) and Bisiach

Figure 3.7. Navon figure. To find the letter H a person has to use global attention, and to find the letter D a person has to use focal attention.

and Luzzati (1978) described patients who not only neglected the left-sided external stimuli but also neglected the left side of internal stimuli or imagery. For example, Bisiach and Luzzati (1978) asked patients primarily who were in a hospital in Milan, Italy, about a famous square in the city of Milan. He told them to image coming out of the front doors of a cathedral situated on one end of this square and to tell him what building on the square they saw in their mind's eye. These patients primarily described the buildings on their right but not on their left. He then asked them to image that they were on the other side of the square looking at the cathedral and to tell him the buildings they now saw. The patients again primarily reported the buildings that would have been on their right, had they been standing across from the church, but these were the same buildings that were initially on their left when they first imagined walking out of the cathedral. The reason I mentioned these reports is that they demonstrate that the brain mechanisms that guide attention to external stimuli might also guide attention to internal stimuli. Thus, when a person images some type of stimulus and uses a global approach, there is a good possibility that this person might be using a right-hemisphere mechanism. If a painter imagines the entire scene before imaging details in the scene and if the scientist images an array of data before he images any specific data, they might be primarily using their right hemisphere.


As I mentioned earlier, Gall thought that if specific anatomic areas mediate specific brain functions then the more brain tissue devoted to this function, the better this function would be performed. Although Gall had posited this about 200 years ago, it was not until about 40 years ago that this hypothesis was scientifically tested. I also mentioned earlier that Wernicke demonstrated that it is the posterior portion of the superior temporal gyrus that contains the memories of how words sound. Thus, if this area of the right hemisphere is injured in people who prefer their right hand, they still will be able to comprehend speech and to speak normally, because the left posterior temporal lobe is intact, but if this region of the brain is injured in the left hemisphere, they will lose their ability to comprehend speech and to speak normally. Geschwind and Levitsky (1968) measured, in the superior portion of the temporal lobe, between the primary auditory area and the end of the Sylvian fissure (planum temporale) on both the right and the left sides of the brain. This area is part of the auditory association cortex. They found that in most people this area, which includes Wernicke's area, was larger on the left side than on the right side (see Figure 3.8). Geschwind and Levitsky's observations supported Gall's hypothesis: Bigger is better. Geschwind and Levitsky, however, did not know the handedness of their participants or if their right hemisphere or left hemisphere was dominant for language. Further support for this anatomic specialization postulate comes from the work of Foundas, Leonard, Gilmore, Fennell, and Heilman (1994), who studied patients who were being considered for epilepsy surgery. Prior to such surgery, patients undergo a procedure called the Wada

Figure 3.8. Diagram of an axial slice of the brain, demonstrating several anatomic areas and the asymmetry of the planum temporale, the left being larger than the right.

test, in which the physicians induce inactivation of first the left hemisphere and then the right hemisphere by injecting a barbiturate into the left and then right carotid arteries (which are the major arteries that feed blood to most of the cerebral cortex). When the language-dominant hemisphere is put asleep, the patient stops speaking. This procedure allows the doctors to know which hemisphere is dominant for speech so that when they remove epileptic brain tissue they can avoid injuring the parts of the brain that are important for speech and language. Foundas et al. (1994) found that language dominance strongly correlates with the size of Wernicke's area or the planum temporale. As I mentioned earlier, when right-handed patients damage the left inferior frontal lobe (Broca's area) they are unable to speak normally because this area is important in programming the muscles that control the articulatory apparatus (e.g., tongue, lips, palate, vocal cords) when making speech sounds. Foundas, Leonard, Gilmore, Fen-nell, and Heilman (1996) also found that the Broca's area on the left (language-dominant side) was larger than that on the right. Right-handed people can make faster and more precise movements with the right hand than with the left hand, and Foundas, Hong, Leonard, and Heilman (1998) also found that in right-handed people the region of the left hemisphere's motor cortex that sends signals that control the right hand is also larger on the left side than on the right side.

If bigger is better, than organisms with bigger brains should be more intelligent than those with smaller brains. Elephants, however, have bigger brains than humans do. Although they have special skills that humans do not have, they do not have the intelligence or creativity of humans. As I mentioned, the brain contains many systems, but only a few of these are important for intelligence and creativity. Although the elephant's brain is larger than the human's brain, it does not contain larger areas of cortex critical for higher order cognitive behavior. People have different sized heads, indicative of different brain sizes. In people with very small heads (microcephaly) there is a high probability for subnormal intelligence and restricted creativity, but there is no evidence that people who have large heads are more creative than those with average-sized heads. Overall size may not be as important as is the size of the multimodal areas in the temporal and parietal lobes.

To be creative a person has to have rich stores of knowledge in her or his chosen field. Detailed knowledge allows one to discover the anomalies that lead to new theories. As I mentioned earlier, one of the most important evolutionary changes in the brain of humans is the development of polymodal and supramodal association areas of the parietal and temporal lobes, including the supramarginal gyrus and the angular gyrus (Brodmann's areas 40 and 39) (see Figure 3.5). I also mentioned that lesion and functional imaging studies have revealed that these areas are important in mediating many higher cognitive activities, such as language, mathematics, and spatial computations. In 1907 E. A. Spitzka studied the brains of several well-known scientists and noted that these eminent mathematicians and physicists had large parietal lobes. Since the time of Isaac Newton, there has been no physicist who was more brilliant and creative than Albert Einstein. Einstein was not a neuroscientist, but he was aware that one of the best means to understand the brain mechanisms underlying creativity was to study the brains of creative individuals. Although Einstein wished to be cremated, he wanted his brain to be used for research. When Einstein died on April 18, 1955, at the Princeton (New Jersey) Hospital, the pathologist at this hospital, Thomas S. Harvey, removed Einstein's brain. Rather than keeping the brain intact, however, so that its overall organization could be studied, Harvey sectioned the brain into 240 blocks and sent different blocks to a variety of different people.

Fortunately, Harvey photographed Einstein's brain before it was sectioned into many little pieces (see Figure 3.9). Witelson, Kigar, and Harvey (1999) viewed these pictures of Einstein's brain and, on the basis of these photographs, they attempted to learn if the brain had aberrant morphology. In primates, including humans, the frontal and parietal lobes are separated from the temporal lobe by a large fissure called the Sylvian fissure (see Figure 3.5). In many people the caudal or posterior end of the Sylvian fissure turns upward, and this upward portion of the Sylvian fissure is called the ascending limb (see Figure 3.5). Normally, this ascending limb separates one of the major portions of the parietal lobe, the supramarginal gyrus, into an anterior and posterior division. When Witelson and her coworkers examined this picture of Einstein's brain, they found that, rather than there being an ascending ramus that divided the supramarginal gyrus into anterior (rostral) and posterior (caudal) divisions at the end of the left Sylvian fissure, the Sylvian fissure in Einstein's brain ended at the postcentral sulcus (see Figure 3.5). On the basis of these observations, Witelson and her colleagues suspected that Einstein had a larger inferior parietal lobe than do most other people, and Einstein's parietal lobe, unlike that of most people, was not divided. They suggested that this large and uninterrupted supramodal cortex allowed Einstein to have a functional advantage in performing mathematics and spatial computations.

Witelson's argument about how the morphology of Einstein's brain might have been responsible for his genius, however, is not

Figure 3.9. Photographs taken in 1955 of five views of Einstein's whole brain (meninges removed): A, superior; B, left lateral; C, right lateral; D, inferior; E, midsagittal view of the left hemisphere. The arrow in each hemisphere (B and C) indicates the posterior ascending branch of the Sylvian fissure as it runs into (is confluent with) the postcentral sulcus. Consequently, there is no parietal operculum in either hemisphere. Scale bar, 1 cm. Reprinted with permission from Elsevier (The Lancet, 1999, vol. 353, p. 2150) and the author, Sandra Witelson.

Figure 3.9. Photographs taken in 1955 of five views of Einstein's whole brain (meninges removed): A, superior; B, left lateral; C, right lateral; D, inferior; E, midsagittal view of the left hemisphere. The arrow in each hemisphere (B and C) indicates the posterior ascending branch of the Sylvian fissure as it runs into (is confluent with) the postcentral sulcus. Consequently, there is no parietal operculum in either hemisphere. Scale bar, 1 cm. Reprinted with permission from Elsevier (The Lancet, 1999, vol. 353, p. 2150) and the author, Sandra Witelson.

consistent with one of the major postulates of neurology. The brain must fit into the skull, and if the skull becomes too large then its size may disable the person who has to carry this burden. As I briefly mentioned, one way of increasing the cerebral cortex without increasing the size of the brain is by developing hills (or gyri), valleys (or sulci), and gorges (or fissures). This gyrification process allows an increase the amount of cerebral cortex to fit into a fixed volume. In an article that was published before the article about Einstein's brain was published, Witelson and Kigar (1992) suggested that the posterior ascending ramus is the continuation of the main stem of the Sylvian fissure and that, unlike gyri, which are formed by the infolding of cortex, the Sylvian fissure—including the ascending ramus—results from uneven growth of the outer cortex relative to inner structure. That Einstein did not have an ascending ramus, therefore, may suggest that the growth of his inferior parietal cortex was not as great as those who do have such a fissure. In addition, whereas creative people such as Einstein are extremely rare, Foundas and her coworkers noted that the posterior ascending gyrus is not present in 15% to 20% of normal people's brains (personal communication, 2002). Thus, an uninterrupted supramarginal gyrus cannot solely account for Einstein's exceptional creativity.

To learn if the size of the parietal lobe might be a critical element in the brains of geniuses, as suggested by Spitzka, I searched the English literature for articles about brain morphology and genius or creativity. I could find no studies that either supported or refuted Spitzka's postulate, and the hypothesis that creative geniuses in a field might have brain morphology that is different from the general population's morphology has not been tested.

Although highly developed portions of the brain can store specialized knowledge and be responsible for exceptional talents, knowledge and talent alone do not allow one to find the thread that unites. Henri Poincare, the famous mathematician, stated,

Among the chosen combinations the most fertile will often be those formed from the elements drawn from domains that are far apart. Not that I mean as sufficing for invention the bringing together of objects as disparate as possible; most combinations so formed would be entirely sterile. But certain among them, very rare, are the most fruitful of all.

In an article my colleagues and I recently wrote about the brain mechanisms of creativity (Heilman, Nadeau, & Beversdorf, 2003), we proposed that creativity may be achieved by applying networks representing internal models in one domain of knowledge to other networks that contain domains of knowledge that share some attributes. We also suggested that many different network architectures probably exist within the association cortices of the brain, raising the possibility that creativity by metaphor might involve the recruitment of networks of substantially different architecture in order to escape the constraints of existing (learned) internal models represented in the networks usually used for thinking in a particular domain.

In an influential article, Mednick (1962) suggested that creativity requires the ability to make associations between separate concepts or ideas. This idea is similar to one of the definitions of creativity described earlier: "finding the thread that unites." Mednick's concept is consistent with many of the tests of creativity that judge a person's ability to make remote associations (e.g., name alternative uses of a brick), and Mednick also developed a test called the Remote Associates Test. In this test the participant is presented with a series of words and the goal for the participant is to recognize how these words are associated (e.g., the words might be blue, American, and cottage, and the correct response might be cheese). The Remote Associates Test for creativity has been validated, and many creative people do seem to perform well on this test. This test, however, primarily assesses verbal-semantic associations, and many forms of creativity require the association of very different forms of knowledge. In J. P. Guilford et al.'s (1978) Alternative Uses Test, the responses of the participant can be scored on several dimensions, such as the number of responses in a fixed time or fluency, the number of different categories to which the named items belongs, and—perhaps most important—originality. For example, in the brick test, using the brick as a doorstop or bookends would be less original than using it for chalk or grinding it up and using it as a cosmetic such as rouge. The anatomic bases of the ability to make the associations needed for creative acts is not entirely known, but the brain's associative pathways could be important.

Corpus Callosum and Interhemispheric Communication

I mentioned earlier that the cognitive functions mediated by the two hemispheres are different. The left hemisphere is dominant for meditating speech and other language functions, such as reading and writing, and the right hemisphere is dominant for spatial computations, including the recognition of faces and the recognition and expression of emotions. In addition to these modular and lateralized networks, the two hemispheres might use different forms of attention and reasoning. For example, the left hemisphere uses focused attention and the right hemisphere use global attention. In general the two types of reasoning used to solve problems are deductive-logical and inductive-probabilistic. Parsons and Osherson (2001) studied these two kinds of reasoning using positron-emission tomography. They found that probabilistic reasoning activated mostly areas in the left hemisphere, whereas deductive reasoning activated primarily regions of the right hemisphere. Deductive-logical and inductive-probabilistic reasoning are both important in the creative process, but these forms of reasoning must be interactive. If creative innovation involves the recruitment and association of networks that have different architecture, different forms of stored knowledge, and different means of thinking and solving problems, then interhemispheric communication and coordination seems to be critical for creative innovation and production. Therefore, the corpus callosum would seem to play a critical role in creativity.

One way of treating medically intractable epilepsy is to disconnect the two hemispheres by cutting the corpus callosum. Lewis (1979) administered the Rorschach Inkblot Test to eight patients who had undergone cerebral commissurotomies (sectioning of the corpus callosum) for intractable epilepsy and noted that disconnection of the two cerebral hemispheres tended to destroy creativity as measured by this test. Bogen and Bogen (1988) noted that there is hemispheric specialization and that although the corpus callosum transfers high-level information, normally this transfer of knowledge or inter-hemispheric communication is incomplete. Bogen and Bogen posited that incomplete interhemispheric communication permits hemispheric independence and lateralized cognition, which is important in the incubation of ideas and perhaps for the storage of independent forms of knowledge (e.g., verbal versus spatial). These authors mentioned Frederic Bremer, who suggested that the corpus callosum serves the highest and most elaborate activities of the brain: in a word, creativity. They also suggest that it is the momentary suspension of this partial independence that accounts for illumination, or what I term creative innovation. They did not say, however, what could account for this momentary suspension of partial independence.

To my knowledge no one has compared the structure and function of the corpus callosum in creative versus noncreative people. If there are differences in callosal communication between people who are matched for brain size, these differences might be functional rather than anatomic, but both of these hypotheses need to be tested.

Intrahemispheric Communication

As I mentioned previously, the cerebral neocortex has six layers. The brain cells or neurons that have a triangular shape are called pyramidal cells. These neurons are primarily found in the third and fifth layers of the cerebral cortex. Some of the axons of the pyramidal cells connect with or synapse onto other nerves in the cortex and hence are called association fibers. Most of the association fibers come from the pyramidal neurons in the third layer in the cortex. There are short association fibers, also called U fibers, that connect neurons in adjacent cerebral gyri. Deeper in the hemisphere, underneath the cortex, there are longer association fibers (axons) that travel in bundles called fasciculi (e.g., arcuate, superior, and inferior longitudinal fasciculi). It is possible that creative people have a richer net of association neurons in layer three or a greater number and variety of associative white matter pathways. Unfortunately, this too has never been systematically studied, but there has been an interesting report about Einstein's brain that suggests that white matter might be important in creativity.

As I mentioned earlier, when Einstein died, the local pathologist removed Einstein's brain and, rather than keeping the brain intact, he sectioned the brain into 240 blocks and sent it to a variety of people. Marion Diamond and her coworkers were among the people who ended up with a piece of Einstein's brain. Little was reported until about 30 years after Einstein's death, when Diamond and her coworkers (Diamond, Scheibel, Murphy, & Harvey, 1985) performed a histological analysis of Brodmann's area 39 (the angular gyrus), where they observed cell types. They found that, compared with the brains of control participants, area 39 on the left side of Einstein's brain contained a higher ratio of glial cells to neurons. These investigators attempted to explain this result by suggesting that this aberrant ratio was "a response by glial cells to greater neuronal metabolic need." This post hoc hypothesis received much criticism but few alternative explanations. Since the work of Dejerine (1891) it has been repeatedly demonstrated that lesions of Brodmann's area 39 (also called the angular gyrus) induce a reading disturbance (alexia). It has been posited that this region stores memories of the visual or written composition of words, and cognitive neuropsychologists have termed these representations the "visual orthographic lexicon." Because Einstein was dyslexic, it is possible that the changes found by Diamond and her coworkers might not be related to an increase of glial cells but rather a reduction of neurons, and it was this developmental abnormality that led to Einstein's dyslexia.

There is, however, an alternative explanation of Diamond and her coworkers' findings. As I mentioned previously, the corpus callosum is primarily composed of myelinated axons whose cell bodies are in the pyramidal layers of the cerebral cortex. The cerebral connectivity important for creativity, however, might be not only interhemispheric (between the right and left hemispheres) but also intrahemispheric (within the hemisphere). In addition to the myelinated axons that carry information between the hemispheres, the myelinated axons in the subcortical white matter carry information between cortical regions in the same hemisphere. These subcortical white matter connections facilitate both inter- and intrahemispheric communication, and inter-hemispheric and intrahemispheric communication might be important for creative innovation because widespread connectivity allows creative people to combine representation of ideas that have been previously isolated. Thus, Diamond and colleagues' findings might suggest that Einstein's parietal lobe had extremely well-developed subcortical white matter, and this finding might suggest that Einstein's creativity was related to his increased interhemispheric and intrahemi-spheric connectivity.

Although in this section I speculate that extremely creative people, such as Einstein, might have more extensive intra- and intercerebral connections, to my knowledge there have been no systematic studies that have tested this hypothesis, but there is another observation that might support the relationship between connectivity and creativity. There are people who, when they perceive a stimulus in one modality, can sense this stimulus in another modality. This phenomenon has been called "synesthesia," which comes from two Greek roots: syn (together) and aesthesia (sensing or perceiving). For example, there are people who, when they hear certain notes or words, see certain colors. This cross-modal sensory excitation can take place between almost any sensory modalities. In addition to cross-modal synesthesia, this phenomenon can also be intramodal such that when people view certain figures in black and white, the figures appear to be in color. If a person without visual intramodal synesthesia were presented with a page that contained the letter L distributed over the entire page and with the letter T inserted between these Ls, and this person had to find the T he or she would have to perform a sequential search attempting to decide if each exemplar was an L or T. If, however, a person has synesthesia, such that when they viewed the letter T it looked red and the Ls looked blue, the letter T would pop out and be found sooner than if all the letters were in black. Ramachandran and Hubbard (2001) examined people who had visual intramodal synesthesia using a similar paradigm and found that the people who had synesthesia recognized the target more rapidly than did control participants. The mechanism that accounts for synesthesia is unknown, but one postulate is that people with synesthesia have greater connectivity between the modules that are interactive (e.g., color and shape). Blakemore and Steven (in preparation, quoted by Underwood, 2003) used functional imaging to study an individual with synesthesia and found that when this person was presented with one type of stimulus (e.g., words) the region that mediates color perception was also activated. According to the connectivity postulate of creativity that I described previously, if people with synesthesia have greater connectivity they should also be more creative. This postulate was supported by the observations of Grossenback (quoted by Underwood, 2003), who found that of 84 people who have synesthesia, 26 had careers that required creativity (artists, writers, and musicians).

Was this article helpful?

0 0

Post a comment