Only quite recently have neuroscientists begun to understand the importance of white matter, a long-neglected part of the brain.

If you set out to understand a building’s lighting system but examined only the fixtures, switches, and outlets, never the wiring in the walls and basement, you might be pretty puzzled. Something like that has happened in the study of the brain, suggests behavioral neurologist Christopher Filley. Scientists now beginning to probe the long-neglected half of the brain called the white matter arediscovering how it specializes in connectivity, with bundles of insulated “wiring” that link neurons within and between gray matter areas into ensembles that may produce the light of conscious mental functioning. A pioneer of research on white matter, Filley proposes a new field of study that would bring “the other half of the brain” into the mainstream of neuroscience. Given the crossword puzzle clue “brain stuff,” what would come to your mind? Quite a few readers might answer “gray matter” and with reason. In common parlance, gray matter has become virtually synonymous with the brain, as when Agatha Christie’s famous detective Hercule Poirot boasts about his “little gray cells.” Much of neuroscience, as presented today, tends to reinforce this view that the gray matter of the brain’s cerebral cortex makes possible our distinctive mental capacities, such as memory, language, thought, and emotion. Yet, a glance at the anatomy of the human brain reveals that cortical gray matter comprises only the brain’s outermost one to four millimeters, a layer about the thickness of heavy cloth, over a brain that in the average human adult weighs three pounds. Almost one half of the brain’s volume is not gray but white matter, the densely packed collection of myelinated (insulated) projections of neurons that course between widely dispersed gray matter areas. If gray matter supposedly “is” the brain, then what is all this white matter doing in our heads? Only quite recently have neuroscientists in the laboratory and clinic begun to understand the importance of this long-neglected part of the brain. In the normal brain, white matter appears to provide the essential connectivity, uniting different regions into networks that perform various mental operations. We know this because, when this connectivity is disrupted by disease or other damage to white matter, the result is often a dramatic disturbance of normal mental function. The scope and variety of syndromes that result from disruption of white matter suggest that white matter makes a pivotal contribution to all realms of human behavior, a contribution we are just beginning to fathom. So diverse and important are these contributions that it is time to consider the need for a new field: the behavioral neurology of white matter. THE DISCOVERY (AND “REDISCOVERY”) OF WHITE MATTER In 1543, in the seventh book of his monumental work, De Humani Corporis Fabrica, the Renaissance anatomist Andreas Vesalius was the first to distinguish clearly between white matter and the gray matter that overlay regions of the cerebral cortex.1 Only with time, though, did the role of white matter in providing structural and functional connections between gray matter areas within the brain become apparent. In the 19th century, the Parisian physician Jean Martin Charcot greatly advanced understanding of white matter’s role with his detailed studies of multiple sclerosis (MS), a disease of young adults characterized by primary damage to the white matter of the central nervous system. At the turn of the 20th century, the ascendancy of Sigmund Freud turned biomedical thinking toward psychoanalytic explanations of behavior, and for more than 50 years all of the brain—white and gray matter alike—was relatively neglected. In 1965, however, Norman Geschwind, M.D., proposed that one mechanism underlying dysfunctional brain-behavior relationships might be cerebral disconnection, a view that pointed directly at the need to study white matter lesions.2 Geschwind’s seminal article advanced the view that the dense connectivity of the brain underlay its mental operations. Central to his concept was the idea that an intricate web of white matter pathways course within and between the brain’s hemispheres. White matter is a vast, intertwining system of neural connections that join all four lobes of the brain (frontal, temporal, parietal, and occipital), and the brain’s emotion center in the limbic system, into the complex brain maps being worked out by neuroscientists. All of the well-known cortical areas such as Broca’s area, Wernicke’s area, the prefrontal cortex, and the hippocampus are connected by white matter tracts to other regions of the brain. This suggests that the cortical regions act in concert to perform mental operations and no cortical area acts in isolation. Without functioning white matter, the brain could be like a group of people in proximity to each other but unable to communicate with each other. After centuries of being recognized but not well understood, white matter gained growing importance as neuroscientists turned to the study of brain-behavior relationships. At the beginning of the 21st century, scientists generally accept the theory that distributed neural networks—widely scattered neurons or clusters of neurons that tend to fire in synchrony—underlie our conscious experiences. This theory has reinforced the idea that the connectivity provided by white matter occupies a central place in the elaboration of human behavior. The impressive growth of modern neuroimaging has let us depict white matter and its functional specializations in increasingly elegant detail. Today, there is an explosion of information on white matter, its role in normal brain function, and its relevance to human illness. EFFICIENT BRAIN SIGNALING In contrast to gray matter, in which the cell bodies of neurons predominate, the term white matter refers to areas of the brain where there is a preponderance of axons coated with myelin. (Axons, which can be up to three feet long, are the longest projections of brain cells and carry a cell’s signal to other cells.) At the microscopic level, white matter consists of millions of closely packed axons, each wrapped concentrically in a myelin sheath, an insulation made up of roughly 70 percent lipid and 30 percent protein. It is this fatty insulation that gives these areas of the brain a whitish hue. Macroscopically, the white matter can be seen to form various kinds of fiber collections, including projection fibers, association fibers, and commissural fibers. Most critical for the higher functions of the brain are the association fibers, which travel within the individual cerebral hemispheres, and the commissural fibers, which connect the left and right hemispheres. Many smaller bunches of fibers, such as the median forebrain bundle, are important as parts of networks devoted to higher functions. In general, however, (with the exception of the corpus callosum—the white matter connecting the left and right hemispheres—and some of the association systems) little is known about which white matter is associated with what types of brain-behavior links. Cells in white matter perfectly support its role in connectivity. Myelin forms a sheath that encircles a cell’s axon in such a way that small regions, known as nodes of Ranvier, are left unwrapped. These unwrapped regions permit an enhanced form of electrical transmission called saltatory conduction. When the cell fires its nerve impulse (or action potential) along its axon, the transmission is much more efficient. This accounts for the much faster conduction of myelinated axons, as compared with axons that are not myelinated. With this special cell physiology, white matter ensures that rapid neuronal conduction takes place in the brain, directly contributing to the efficiency of information processing typical of normal cognition. Not surprisingly, an abundance of white matter is a striking feature of the frontal regions of the brain. These lobes have the highest degree of connectivity of any brain lobe. Our frontal lobes are in a unique position, by way of their white matter tracts, to serve as integrators of mental operations performed by other cerebral regions. WHITE MATTER ACROSS THE BRAIN’S LIFE SPAN White matter develops and changes across our life span in a pattern strikingly different from gray matter. We do not complete the formation of brain myelin until many years after birth, perhaps past 20 years of age; by contrast, we have our full complement of brain neurons at birth. In later life, however, a slow but steady loss of white matter occurs that may be greater than the loss of neurons (which is now thought to be less pronounced than formerly believed). So it appears that there is less white than gray matter at both ends of the human life span. Provocative clinical implications of this pattern are emerging. In children and adolescents, for example, the maturation of white matter, particularly in the frontal lobes, may correlate with the acquisition of mature aspects of personality such as motivation, demeanor or bearing, and executive function. These attributes, long associated with the frontal lobes, are among the last to develop, sometimes not fully maturing until our early twenties. This development may parallel the timing of full completion of the myelination of white matter tracts that connect the frontal lobes to other brain regions.3 At life’s other pole, the aging brain is characterized by a selective diminution of cerebral white matter. In an autopsy study of normal brains from age 20 to 90, white matter loss, by volume, was 28 percent, but neocortical (gray matter) loss, was 12 percent, less than half of white matter loss.4 This greater loss of white matter, with its special role in connectivity and efficient brain communication, suggests a cause for slowed speed in information processing, diminished attentional capacity, and forgetfulness—some of the typical cognitive changes of aging. Normal age-related loss of white matter may contribute to mild cognitive impairment, although the concept of such impairment is still controversial. According to one recent, typical proposal, it marks a transition from normal aging to dementia. Although mild cognitive impairment is most often viewed as a precursor to Alzheimer’s disease, the changes in the brain that cause the impairment are still not known. Possibly the cause in some individuals will be found in white matter. Even if mild cognitive impairment turns out to be a harbinger of Alzheimer’s in some people, age-related changes in white matter have a wider significance for the cognitive functioning of older people, particularly when normal age-related changes in myelin are exacerbated by disease or injury in white matter. NEUROIMAGING WHITE MATTER Although introduction of computerized axial tomography (the CAT scan) in the 1970s revolutionized how researchers could visualize and study the brain, the detailed interpretation of the structure of white matter awaited the advent of magnetic resonance imaging (MRI) in the 1980s. MRI became the method of choice not only for research on white matter and its disorders but for diagnosis of MS and many other problems. Older disorders were better understood, new ones recognized. It became possible to correlate a patient’s complex behavioral problems, particularly in the case of higher mental functions, with white matter pathology. Still newer imaging technology has further refined our visualization of white matter. Most exciting is diffusion tensor imaging, which measures the dispersion of water within white matter tracts. In normal white matter, water diffuses in the direction of the specific tract being imaged (called anisotropic diffusion). In damaged white matter, water diffusion is isotropic, meaning it is less directional and more chaotic. Because diffusion tensor imaging can detect these different types of diffusion, it offers the intriguing prospect of mapping the configuration and connectivity of both abnormal and normal white matter regions. Using magnetic resonance spectroscopy, we can discover the chemical composition of white matter regions. Sometimes called a noninvasive biopsy, magnetic resonance spectroscopy can be used to measure certain products of cell energy use in selected regions, giving us a clue to the integrity of the myelin and axons. Evidence is growing that this technique can find white matter alterations even in areas that appear normal on conventional MRI. Finally, magnetization transfer imaging (a technology based on the interactions of protons and macromolecules) generates data on the fine structure of white matter, yielding still more information on myelin integrity and possible myelin injury, including in otherwise normal-appearing white matter. Such neuroimaging techniques reveal structure, but there are equally impressive imaging studies of brain function. To date, these studies have focused primarily on the cerebral cortex because of gray matter’s obvious role in mental operations and its high use of energy, which can be detected by positron emission tomography (PET). A less expensive but also less elegant technique of this type is single photon emission computed tomography (SPECT). More recently, functional MRI (fMRI) was added to the available imaging methods. The neuroimaging of function and structure are complementary in pursuing understanding of higher brain capacities. In particular, PET and fMRI can identify cortical regions involved in cognitive processing, whereas other MRI methods can establish the patterns of connections between these areas. Most scientists are persuaded that distributed neural networks, linked into arrays of both gray and white matter structures, may underlie our conscious mental operations. It should be possible to map these neural networks and link them to specific functions, such as memory and attention, by combining structural and functional neuroimaging methods. This exciting prospect is driving new research efforts. WHITE MATTER DISORDERS The white matter is vulnerable to diseases and injuries with widely varying causes, processes, locations, degrees of seriousness, and courses of development. Prognosis and treatment constantly challenge neurologists. Viewing the disorders as a group, however, we can identify common themes and find new clues to cognition. Keep in mind that, in some disorders primarily of white matter, there are effects in the gray matter (and vice versa)—hardly surprising. In MS, for example, some demyelination occurs in cortical gray matter areas; in Alzheimer’s, a classic gray matter disease, damage to white matter tracts is most likely caused by loss of the axons of neurons in the cerebral cortex that have died. Does this complexity mean that white matter disorders per se are not worth studying? Hardly. But we must recall that effective study of disorders in neurology, psychiatry, and all branches of medicine requires recognition that some aspects of the disorder will fall into other disciplines. We can learn as much by examining details of various white matter disorders as by pursuing fine distinctions among degenerative diseases of the cortex, such as Alzheimer’s and frontotemporal dementia, or the subtleties of bipolar disorder in contrast to schizophrenia. The focus on white matter serves to introduce a host of testable hypotheses and guide research on distributed neural networks involving systems of gray and white matter structures working together. Cerebral white matter disorders can be genetic, demyelinative, infectious, inflammatory, toxic, metabolic, vascular, traumatic, neoplastic, and hydrocephalic. Each classification signifies a distinct disease process, and, within classifications, diseases vary greatly. What, then, are some commonalities among the more than 100 white matter disorders in how they affect brain and behavior? To begin with, all are associated with cognitive or emotional dysfunction of some kind and similarities in brain-behavior dysfunction cut across disease categories. In infants and children, genetic diseases such as metachromatic leukodystrophy involve the failure of brain myelin to develop normally. Subsequent dysmyelination leads to early disability and death, or, occasionally, a common sequence of psychosis followed by dementia. Young adults are at high risk of the demyelinative diseases such as MS, whereby inflammatory destruction of myelin leads to neurologic and neurobehavioral disability. Some infectious diseases that induce cognitive decline do so, in part, by affecting the white matter. Prominent among these diseases is the acquired immune deficiency syndrome (AIDS) dementia complex. Similarly, noninfectious inflammatory diseases such as systemic lupus erythematosus can affect the white matter by immune-related processes, and the multiple manifestations of neuropsychiatric lupus are increasingly recognized. A large and growing number of white matter toxins have been identified, among them the common industrial and household solvent toluene. Detailed MRI studies of toluene leukoencephalopathy in solvent abusers disclosed dramatic white matter disease that correlates with the severity of dementia produced. Metabolic disorders of white matter also exist, including dementia from vitamin B12 deficiency. One of the most common white matter disorders is a variant of vascular dementia called Binswanger’s disease. Traumatic brain injury also qualifies as a white matter disorder, because patients with brain trauma often have a white matter lesion known as diffuse axonal injury. Some brain cancers affect primarily white matter. Finally, hydrocephalus (fluid on the brain) exerts its major effect on the white matter around the lateral ventricle of the brain, and examples in children (early hydrocephalus) and adults (normal-pressure hydrocephalus) both indicate that the resulting white matter damage can have serious consequences for brain and behavior. GOOD NEWS AHEAD? Given this diversity, we cannot generalize about prognosis and treatment. Also, any prognosis must consider not only the illness but how much damage it has already done, whether the illness is chronic, how old the patient is, and any coexisting problems. Nevertheless, certain guidelines apply, and they are revealing. One important factor is the stage of the disease since, in white matter disorders, there is commonly a range of manifestations from mild and reversible pathology to severe and irreversible tissue death. This variability is most evident in the toxic leukoencephalopathies but likely applies to other white matter disorders, as well. In particular, evidence is good that when there is extensive loss of axons, the prognosis is worse in white matter disorders such as the leukodystrophies, MS, AIDS dementia complex, and Binswanger’s disease. Because white matter disorders can be mild and reversible, however, the prognosis is more optimistic than for gray matter disorders, most notably the dreaded dementia of Alzheimer’s disease in which cortical cell bodies, synapses, and receptors are destroyed by the disease. In many white matter cases, such as subtle toxic leukoencephalopathies characterized only by myelin swelling, full recovery is expected after withdrawal of the toxin, because the myelin and axons have not been permanently damaged. Even when they are damaged, substantial recovery is still expected if axons are not destroyed. For physicians, this means that both prevention and early treatment offer high potential for helping their patients. Prevention applies especially to white matter disorders that are infectious, metabolic, toxic, vascular, and traumatic. Early treatment can retard, arrest, or reverse the disease process in many white matter disorders before the loss of axons. As medical and surgical treatments for white matter disorders improve, opportunities for effective early intervention do also. Treatment of MS with drugs that modulate immunity, such as interferon-beta-1a, appears to have a beneficial effect on the patient’s cognition. Exciting possibilities are on the horizon for many white matter disorders using gene therapy and stem cells. Potentially, stimulants and cholinesterase inhibitors will counteract the slowing of attention and memory loss. WHITE MATTER DEMENTIA Cognitive dysfunction is the most common brain-behavior syndrome related to white matter disorders, and, in many cases, the dysfunction is severe enough to be called “white matter dementia.” The prevalence of cognitive dysfunction and dementia from white matter disorders as a whole is uncertain; the epidemiologic data are just not available. But much clinical experience supports the sense that cognitive dysfunction is prevalent among these white matter disorders. Moreover, research suggests that syndromes of widespread white matter dysfunction far outnumber syndromes of isolated, regional white matter dysfunction. A case in point is MS, in which cognitive dysfunction or dementia may afflict as many as two thirds of patients, whereas more specific problems such as language impairment, known as aphasia, occurs in less than one percent of patients. Similarly, although neuropsychiatric syndromes such as depression are common in patients with white matter disorders, they may result from many causes, so the cause-effect relationship is less clear. All this suggests that cognitive impairment will prove to be a leading source of clinical distress and disability in cases of damage to cerebral white matter. Not surprisingly, in early stages of any white matter disorder milder cognitive dysfunction is more common than dementia, but dementia often follows. In MS, for example, estimates are that 10 to 20 percent of patients will develop dementia. Despite this, it is important to realize that in clinical practice the recognition of early cognitive dysfunction in the white matter disorders is far from simple. Many patients show subtle cognitive symptoms and signs, frequently co-mingled with other neurologic or medical features of their disease, challenging the clinician to interpret the relationship of white matter manifestations to cognitive status. Moreover, the range of clinical features that herald the onset of cerebral white matter involvement is impressively broad: inattention, executive dysfunction, confusion, memory loss, personality change, depression, somnolence, lassitude, and fatigue. This nonspecific clinical profile often suggests a primary psychiatric disorder, and many patients with white matter dementia do display early psychiatric dysfunction before measurable cognitive impairment. Here, we see the relevance of white matter disorders to the growing field of neuropsychiatry, but we still have no commonly accepted term to describe white matter disorders presenting as early cognitive impairment. One suggestion is to use the term “dysmentia,” meaning disordered (Greek dys) mind (Latin mens). Properly defined and standardized, dysmentia could describe early cognitive impairment in white matter disorders, just as mild cognitive impairment describes early cognitive loss preceding the development of the gray matter disease recognized as Alzheimer’s. How do cognitive dysfunction and dementia actually present themselves to the physician? What profile of deficits and strengths can we use in diagnosis, counseling, rehabilitation, and research on new therapeutic strategies? A profile seems to be developing that includes a sustained attention deficit, executive dysfunction, memory retrieval deficit, visuospatial impairment, and psychiatric dysfunction with normal language, motor function, and procedural memory. Although still preliminary, this specific combination differs from that seen with cortical dementia such as Alzheimer’s disease and from subcortical gray matter dementias such as Huntington’s disease. Again, we see cerebral white matter’s possible unique role in the organization of cognition and emotion. Sustained attention deficits, executive dysfunction, and memory retrieval deficits are most typical of patients with white matter disorders; all relate to a general slowing of cognition, often called impaired speed of information processing. Sustained attention deficits, executive dysfunction, and memory retrieval deficits are most typical of patients with white matter disorders; all relate to a general slowing of cognition, often called impaired speed of information processing. In terms of brain anatomy, sustained attention (concentration, vigilance), executive thinking, and memory retrieval are all closely associated with the operation of the frontal lobes, and most white matter disorders show a preference for the frontal white matter. Even when white matter lesions are situated in more posterior cerebral locations, frontal lobe functions are still affected, probably because of the dense connectivity between frontal and other regions. Visuospatial skills are also affected in white matter disorders. In contrast, language is typically normal or only mildly affected in patients with white matter disorders because the language-related cortex is spared. Motor function also tends to be intact, in keeping with the relative sparing of deep gray matter structures. Likewise, procedural memory, or memory for skills such as bicycle riding, is retained. What about white matter lesions linked with narrower brain-behavior disturbances, including classic syndromes such as aphasia and amnesia? Although these are rightly viewed as more common with cortical lesions, recent research also links them with white matter damage. For example, there are reports of isolated amnesia associated with stroke that affects a white matter region called the mamillothalamic tract. A language disturbance known as conduction aphasia is related to MS plaque in another part of the brain, the left arcuate fasciculus. Thus, although they are uncommon compared with syndromes caused by diffuse white matter damage, the focal brain-behavior syndromes illustrate the importance of white matter tracts in all domains of higher function. Research here can enhance our understanding of the neural networks that underlie these higher brain functions. WHITE MATTER AND NEUROPSYCHIATRY Abnormalities of cerebral white matter are associated with a spectrum of emotional disturbances. This category of disorders is vaguer than the brain-behavior syndromes because the correlation of white matter disorders with psychiatric syndromes is much less clear; and psychiatric impairments are notorious for having multiple causes. Still, there is much new information on the role of white matter in emotional function, shedding light on both white matter disorders and psychiatric diseases. These neuropsychiatric syndromes fall into two general groups: psychiatric features in patients who have known white-matter disorders, and the many psychiatric diseases in which white matter abnormalities are implicated. In patients with known white matter disorders, reports document the presence of depression, mania, psychosis, pathologic crying or laughing, and euphoria. We do not know, as yet, how closely these psychiatric syndromes correlate with measures of white matter dysfunction. Thus, the possibility remains that a given psychiatric syndrome is related only indirectly, or not at all, to the patient’s specific white matter disorder. When it comes to primary psychiatric diseases, often considered idiopathic (of uncertain cause) and as yet not linked with structural brain damage, there are recent intriguing reports from neuroimaging research on the structure of white matter. In patients with schizophrenia, for example, imaging studies have detected microscopic abnormalities in white matter structures, and widespread myelin and oligodendrocyte dysfunction are linked with altered cerebral connectivity.5 Much evidence also supports an association between white matter changes and geriatric depression, although a firm correlation has yet to be established. MRI studies have found that white matter abnormalities are more common in patients with bipolar disorder than in the general population. In children with attention deficit/hyperactivity disorder, a diminished volume of right frontal white matter was found to correlate with impaired sustained attention. In contrast, an increase in the volume of hemispheric white matter in all lobes was observed in autism. Finally, diffusion tensor imaging studies of schizophrenic men found a correlation of inferior frontal lobe white matter abnormalities with impulsive aggression. Obviously, we need far more detailed investigation of how white matter abnormalities may contribute to psychiatric disease, perhaps by disrupting neural networks devoted to emotional function. WANTED: A BEHAVIORAL NEUROLOGY OF WHITE MATTER The study of higher functions in humans requires consideration of all the brain’s neural tissues. Long neglected as a contributor to the organization of cognitive and emotional operations, white matter is the object of intense, intriguing, and increasingly fruitful efforts to improve our understanding. Studying people with white matter disorders to correlate their brain lesions with specific behavior changes promises a wealth of insights. Increasingly, this method will be complemented by sophisticated neuroimaging techniques that yield detailed visualization of white matter tracts as they participate in the cognitive and emotional operations of distributed neural networks. In practical terms, an appreciation of the brain-behavioral importance of white matter disorders can greatly benefit patients, especially as early recognition and treatment often determine an outcome. In theoretical terms, further study of white matter and its disorders expands our knowledge of the brain as an extraordinarily complex structure in which the connectivity provided by white matter is central to cognition, emotion, and consciousness itself. In the most general sense, the gray matter of the brain facilitates information processing, and the white matter facilitates information transfer; both are critical for efficient operation of the neural networks responsible for a specific mental domain. In the presence of damaged white matter, information processing occurs only in a slowed and inefficient manner, and, if the white matter is severely impaired, there may be no processing at all. Considerations like these argue strongly for the evolving field of the behavioral neurology of white matter, an organizing framework that can stimulate urgently needed study, with no less a goal than a more complete and powerful portrait of the organ of the mind. References: 1. Saunders, JBDM, and O’Malley, CD. Illustrations from the works of Andreas Vesalius of Brussels. New York. Dover, 1973. 2. Geschwind, N. “Disconnexion syndromes in animals and man.” Brain 1965; 88: 237-294, 585-644. 3. Olesen, PJ, Nagy, Z, Westerberg, H, et al. “Combined analysis of DTI and fMRI data reveals a joint maturation of white and grey matter in a fronto-parietal network.” Cognitive Brain Research 2003; 18: 48-57. 4. Pakkenberg, B, and Gundersen, HJG. “Neocortical neuron number in humans: effect of sex and age.” Journal of Comparative Neurology 1997; 384: 312-320. 5. Davis, KL, Stewart, DG, Friedman, JI, et al. “White matter changes in schizophrenia. Evidence for myelin-related dysfunction.” Archives of General Psychiatry 2003; 60: 443-456.