Cerebrum Article

Cognitive Fitness for the Older and Wiser

Neuropsychologist Elkhonon Goldberg’s book The Wisdom Paradox (excerpted in the Spring 2005 issue of Cerebrum) described his new theory of how brain changes in older age may be surprisingly positive, undergirding an increasing cognitive mastery. Here, Goldberg explores how recent neuroscience research has begun to overturn some long-held assumptions, supporting his hypothesis that a marked shift in hemispheric specialization as humans age takes advantage of the brain’s plasticity to reward lifelong cognitive exertion with the gift of wisdom.

Published: October 1, 2005

People may like to imagine science over the course of centuries as a steady, triumphant march into the unknown, advancing the boundaries of knowledge. Fair enough, but shorten the time frame to decades, and science at times resembles women’s fashions, dominated by ever-changing trends, all the rage today, sidelined tomorrow.

Consider a striking example. Until relatively recently, the idea of cognitive development in older age was a fantasy. Neuroscientists had defined cognitive development as applying strictly to the transition from infancy to maturity. Then, in just a decade or two, cognitive changes attending the later part of life soared to the top of the neuroscience research agenda. As just one indication of this change, I would point to the publication of the Cognitive Neuroscience of Aging (Oxford University Press, 2004), a comprehensive multiauthor overview of this new field. In the same vein, the capacity of the brain to change physically, including the ability to create new neurons, was once understood to occur exclusively during early development. Its application to the adult brain was regarded, at best, as marginal, even fanciful. With the exception of an iconoclastic few, most scientists denied the possibility of new brain cells in adulthood. But today this ability, an aspect of neuroplasticity, and its existence as a lifelong phenomenon are among the hottest areas of neuroscience.

By contrast, the question of hemispheric specialization—the respective roles of the brain’s right and left halves—was a mainstay of neuropsychology and cognitive neuroscience in the 1970s and early 1980s, with a record number of journal articles, conference presentations, and doctoral dissertations on the subject every year. Then, sometime in the late 1980s, interest in the topic waned, as though the puzzle of hemispheric specialization had been fully solved or it had been suddenly dismissed as inconsequential. In fact, neither was the case. Hemispheric duality remains arguably the cardinal and most striking feature of brain form and structure, but scientists still do not fully understand the corresponding functionality. The decades-long quest to understand hemispheric specialization is an example of how Hegel’s thesis—at a certain point sufficient quantity makes a leap to quality— fails to materialize. Volumes of published research yielded relatively little additional fundamental insight relative to what was already known by the end of the 19th century. Scientists, with a healthy measure of intellectual opportunism, moved on and away from the intractable puzzle of hemispheric specialization in exasperation rather than in exhilaration, to graze on more welcoming pastures.

Yet, despite opposite trajectories in the grant-driven race of scientific fads, an intricate relation exists among cognitive aging, neuroplasticity, and hemispheric specialization. By considering them together, scientists inform and enrich each and arrive at an unexpected synthesis.

Exploding Right-Brain/Left-Brain Dogmas

Why has understanding of the respective roles of the brain hemispheres eluded scientists for so long? One reason, I believe, is that scientists have assumed that these roles are static. It has been assumed that the so-called dominant (usually left) hemisphere is in charge of language, and the nondominant (usually right) hemisphere is in charge of nonverbal, mostly visual-spatial, matters. Even alternative, more imaginative, theories of hemispheric specialization are derived from this fundamental distinction. Some theories, for example, link the left hemisphere to analytic processing and the right hemisphere to holistic processing, or link the left hemisphere to conscious processing and the right hemisphere to nonconscious processing.

Embracing the language versus visual-spatial dichotomy of hemispheric specialization in its narrow, literal form has led to a dead end. First, scientists have not used the link between language and the left hemisphere to help them understand the wider evolutionary continuities that culminated in the emergence of language, but which surely began with something else. Instead, they accepted a kind of evolutionary deus ex machinae—the appearance of two important traits in humans—hemispheric specialization and capacity for language—without any clear evolutionary antecedents. Now this error is being exposed by new evidence that functional hemispheric specialization is present in nonhuman primates—obviously in the absence of language, at least as narrowly defined.

We also know something today that was not apparent when neuropsychology first embraced the verbal-nonverbal dichotomy: namely, that overall differences in form and structure, cellular composition, and biochemistry exist between the hemispheres and, most importantly, that these differences are not limited to humans but are shared by certain other mammals.

Much of the research comparing the form and structure of the hemispheres was pioneered by Norman Geschwind, M.D., and his associates in the late 1960s and 1970s. They and others discovered important differences that include, among others, a thicker cortex in the right hemisphere than in the left hemisphere and a different shape of the sylvian fissure, which has a higher posterior end in the right hemisphere.

On the cellular level, the differences include a kind of brain cell known as the spindle cell, with exceptionally long axons, which is more numerous in the right hemisphere. On the biochemical level, differences were discovered in the activity of various neurotransmitters, including more vigorous expression related to norepinephrine in the right hemisphere and dopamine in the left hemisphere, as well as an unequal representation of estrogen receptors. These differences were first studied by Stanley Glick, M.D., Ph.D., and his colleagues in the 1980s and are being actively pursued by researchers today.

We are also beginning to learn about the genetic basis of some of these asymmetries. Tao Sun, Ph.D., and his colleagues at Harvard Medical School recently studied the differing levels of gene expression between left and right hemispheres in mouse embryos. Their research, reported early in 2005, raises the possibility that human left-right brain specialization reflects asymmetric cortical development at early stages of life.1 We are also learning how aberrant forms of these asymmetries may play a role in various neurodevelopmental disorders. In one intriguing example, Martha Herbert, Ph.D., and her colleagues at the Center for Morphometric Analysis at Massachusetts General Hospital recently published results of a whole-brain magnetic resonance imaging (MRI) survey that showed differences in asymmetry between children with autism and developmental language disorders and normal children, these differences being most significant in the higher-order cortical areas.2

Knowledge of these asymmetries in humans alone should have given the devotees of the verbal-nonverbal dichotomy serious pause. True enough, one can easily see how the left-hemispheric dominance for language can be linked to some of these differences. For instance, the planum temporale and frontal operculum, which are larger in the left hemisphere, are both involved in language. But it would take an extreme feat of mental gymnastics to explain how some of the other structural and biochemical differences that have been discovered also relate to the verbal-nonverbal dichotomy. To the extent that scientists believe in a relation between neural structure and neural function, the existence of such differences must imply corresponding differences in the functions of the human cerebral hemispheres, beyond a simple verbal-nonverbal dichotomy, but, as yet, they do not know what these differences are.

An even larger dilemma looms. Beginning with the observations of cerebral asymmetries in nonhuman primates by Marjorie LeMay, M.D., and her colleagues in the 1970s, it gradually became clear that virtually none of the asymmetries between the hemispheres mentioned earlier is unique to humans. Nearly all are present in other primates, and many are present even in nonprimate mammals. Even the gender differences evident in human hemispheric asymmetries (which are more pronounced in males than in females) are foreshadowed in other species. In addition, none of the nonhuman mammalian species possesses language in its narrow definition. If hemispheric differences exist at various stages of mammalian evolution, then it stands to reason that functional differences between the hemispheres also exist at various stages of mammalian evolution, and that they themselves have changed over time.

The Known and the Novel

Although clearly appreciated by some neuroscientists, the evolutionary perspective has not been embraced by most of the field. Hemispheric specialization continues to be implicitly viewed as a sudden and mysterious appearance of a trait, which, once in place, becomes static, not affected by change.

The tendency to cling to a static view of hemispheric specialization has affected thinking not only on the time scale of mammalian evolution but also of the individual human life and even of the acquisition of specific cognitive skills. The static view often takes the form of a rigid neuropsychological list that assigns certain cognitive functions to the left or right cerebral hemisphere in an absolute, inviolate fashion, with little room for individual or group differences, or for change over time. Thus, hemispheric specialization is simply assumed to be alike, even identical, in the young and old.

According to this notion, language and skilled praxis reside in the left hemisphere, whereas facial recognition resides in the right hemisphere, as does the processing of music. Unfortunately this view, whatever its appeal, is wrong. It is time to give the question of hemispheric specialization a fresh look in a more dynamic framework.

In my own quest to find a suitable framework to capture the dynamic nature of hemispheric specialization (and also integration), I became increasingly convinced that the right hemisphere is superior at dealing with cognitively novel situations, and the left hemisphere is superior at implementing well-entrenched cognitive routines. This way of looking at hemispheric specialization does not, for the most part, negate the association between language and the left hemisphere. It merely subsumes it as a special human example of a more fundamental principle, which operates across a range of mammalian species. I refer to this principle as the novelty-routinization principle of hemispheric specialization.

The novelty-routinization hypothesis is radically different from the earlier theories in several important respects. First, it can be applied to any species possessing a hemispheric brain capable of learning. This, in turn, allows us to explore the evolutionary origins of hemispheric specialization in a way not afforded by the more traditional verbal-nonverbal distinction, which is limited to humans. Second, the novelty-routinization hypothesis rejects the idea of an intrinsic, fixed link between the type of mental activity and the side of the brain (for example, an obligatory association between facial recognition and the right hemisphere). According to the novelty-routinization principle, a cognitive task’s predominant association with one side of the brain is determined more by the task’s position on the novelty-familiarity continuum and less by the task’s intrinsic properties. Novelty, of course, is a relative notion. What is novel for Joe Blow may be quite familiar to Jane Blane, and vice versa. Therefore, the novelty-routinization principle implies a far greater emphasis on individual differences in the lateralization of various mental activities. Finally, and most important for our present discussion, the novelty-routinization hypothesis highlights the changing nature of how these activities are handled by the brain during a person’s life. What is novel for that person today will be familiar tomorrow, or in a week, or in a year. What is familiar today was novel yesterday, or last week, or last year.

The transition from novelty to familiarity is the essence of learning and is fundamental to human cognition, to myriad specific cognitive skills at different stages of formation. My thesis is that this process is driven by a constant neural dynamic that shifts cognitive control of specific skills from the right cerebral hemisphere to the left cerebral hemisphere and that this shift is what it means to learn a skill. If this thesis is correct, the dynamic hemispheric interaction in learning should be demonstrable on various time scales for various situations, ranging from the time scale of hours or days for experimental cognitive skill learning in the laboratory, to years or perhaps even decades for various forms of real-life cognitive proficiencies and professional expertise.

This is exactly what has been shown in experiments that test visual and auditory perception, memory, and learning. The low-tech methods of the 1970s and 1980s used an apparatus called a tachistoscope that flashed images on a screen rapidly to test the visual component and headphones that conducted different sounds separately and independently into each ear to test the auditory component.3, 4 More recent studies use functional neuroimaging, such as positron emission tomography (PET) and fMRI scanning, to capture the hemispheric interaction that occurs in learning. During the early stages of a new cognitive task, as long as the task is still novel and the study participants are still relatively task naive, the right hemisphere plays an important role. As the participants become increasingly familiar with the task and proficient at its execution, the scans show that the left hemisphere takes over. The novelty-familiarity aspect of the task seems to override all the others, including whether the task is verbal or spatial or whether it involves an artificial laboratory skill (for example, manipulating a robotic arm), which takes a few hours to learn, or a real-life complex competency that takes years or decades to form, such as mastering a musical instrument.

Does a Hemispheric Shift Underlie Wisdom?

Does an understanding of hemispheric specialization as dynamic, not static, change our understanding of the aging brain? Throughout our lives, we accumulate an ever-increasing repertoire of cognitive skills, competencies, and capacities for recognizing familiar patterns that enable us to deal with new situations as though they were familiar because we recognize an underlying pattern. As we age, our mental life becomes increasingly dominated by cognitive routines, or mental “autopilots,” which is not necessarily a bad thing. The gradual accumulation of such cognitive assets is a unique fruit of aging, an essential aspect of what we call wisdom. A well-stocked arsenal of cognitive routines enables an aging mind to solve complex problems by using the effortless and near-instantaneous mechanism of pattern recognition. The same problems would likely pose a stiff challenge to a much younger mind, perhaps endowed with sharper memory, nimbler reaction time, and more focused attention, but lacking an arsenal of ready-to-use cognitive routines.

But if the right hemisphere is in charge of novelty and the left hemisphere is in charge of mental routines, then, as such routines accumulate, and as our reliance on them increases, our dependence on the left hemisphere must gradually increase and our reliance on the right hemisphere must gradually decrease. One would expect this gradual shift to be a lifelong trend, extending far beyond the time frame of language acquisition and reaching well into advanced age. This idea, first proposed by Jason Brown, M.D., and Joseph Jaffe, M.D., in 1975, and by myself and Louis Casta, Ph.D., in 1981, awaited confirmation for a few decades.

Today, such confirmation comes from two areas of intense neuroscientific interest: the effects of aging on the brain and neuroplasticity. As it turns out, the unwelcome changes that attend aging do not affect all of the brain equally. One such disparity came to the attention of neuroscientists only recently: The right hemisphere shrinks more rapidly than the left hemisphere as we enter the advanced decades of our lives. This uneven or asymmetric atrophy was shown in a study conducted by scientists at Johns Hopkins University and the National Institute on Aging and presented at the meeting of the Organization for Human Brain Mapping in 2003. Researchers found that, as we age, the grooves on the surface of the brain (cortical sulci) become shallower in the parietal and occipital regions of the right hemisphere. This suggests a deterioration of the elevated ridges (gyri) that surround the sulci. By contrast, the left hemisphere seems to be less affected by this process.

At the same meeting, a group of Japanese neuroscientists reported similar results from an experiment that assessed the three-dimensional size of structures on the cortical surface by using brain imaging to measure brain volume. Their analysis also showed that the decline in the size of gyri becomes evident at least a decade earlier in the right hemisphere (the fourth decade of life) than in the left hemisphere (the fifth decade of life). A group of Australian scientists led by Evian Gordon, M.D., measured other specific regions such as the insula, which produces an emotionally relevant context for sensory experiences. The insula was shown to have similar asymmetrical atrophy that affects the left hemisphere less than the right.

The Astounding Plastic Brain

The flip side of observing deterioration in the right hemisphere would be finding factors that, at least to some degree, protect the left hemisphere from the decaying effects of aging. Enter neuroplasticity. As mentioned earlier, scientists long thought that neurons stopped proliferating in the brain early in life and that, from then on, it was all downhill: Neurons were steadily lost in the tens of millions without being replaced. A series of spectacular studies conducted in recent decades overturned this grim dogma. New neurons continue to be born in the brain throughout our whole lives, even when we are old. The first evidence for this came from animal experiments by Fernando Nottebohm, Ph.D., Elizabeth Gould, Ph.D., and others during the final years of the 20th century. Human evidence quickly followed with the seminal report of Swedish scientist Peter Eriksson, M.D., and his colleagues who showed that the human hippocampus retains its ability to generate new neurons throughout life.5

What is even more astonishing and significant is that this lifelong capacity for neuronal proliferation, to a large extent, depends on one’s mental activity. A vigorous mental life stimulates vigorous proliferation of new neurons. The analogy between the effects of physical exertion on the muscles and cognitive exertion on the brain is unmistakable.

Evidence that the human brain retains its ability to produce new neurons was followed with more spectacular results that most neuroscientists did not expect even a few years ago. It turns out that the effects of cognitive activation on neuroplasticity are large enough to be observed in the actual size of the brain structures. These discoveries appeared during the first years of the new millennium and have drawn considerable attention, far beyond scientific journals.

The first discovery came from the now-famous London cabdrivers, who have larger hippocampi relative to the rest of the population. Moreover, the longer they have been on the job, the larger their hippocampi. The hippocampus is the sea-horse shaped part of the brain known to be particularly important in spatial memory. Who strains his or her spatial memory more than a cab driver traversing a huge metropolis from one end to another, day in, day out?

This was the first evidence to show that a particular kind of mental activity enhances the growth of certain parts of the brain, even in adults. But it is certainly not the only evidence. It turns out that an area of the cortex singularly important for processing music—the Heschl’s gyrus—is particularly large in musicians. As is the case with the hippocampi of cabdrivers, the size of the Heschl’s gyrus is positively correlated with the number of years spent in practice. Again, the effects of neuroplasticity seem to override the age-associated decay of brain tissue. Likewise, the brain structure particularly important in language, the left angular gyrus, is larger in people who are bilingual than in those who speak only one language, and this effect is not limited to young people; it is also present in those who (like me) acquired a second language later in life.

These discoveries are remarkable in several respects. First, new neurons continue to proliferate over the life span in humans, not just in animals. Second, the loss of neurons with age can be delayed, and delayed dramatically, by mental exertion. The effect must indeed be dramatic if the actual size of vital brain structures is observed to be increased, as was the case in the cabdrivers, bilingual speakers, and musicians. Third, the effects are relatively specific; new neurons multiply in certain parts of the brain with particular vigor depending on the nature of mental activity.

Cognitive Fitness: The Benefits and Limits

These research conclusions are ripe with not only theoretical but also practical implications. To a neuroscientist, they help explain why the left hemisphere appears to hold its own in the face of aging better than the right hemisphere. Because our pattern-recognition arsenal continues to accumulate in the left hemisphere, we increasingly rely on it and exercise it more. By contrast, the right hemisphere, so keen at facing novel cognitive challenges, is used less as we age, because an increasing number of cognitive challenges cease being novel and become familiar.

For all of us, these discoveries highlight the importance of vigorous mental activity as we age. Just as physical exercise confers health benefits for the cardiovascular system and serves as a protective factor against heart attack, cognitive exercise confers certain health benefits for the brain and serves as a protective factor against cognitive losses associated with normal aging, and possibly even, to a degree, against damage related to dementia.

We can also conclude that cognitive exertion should be diverse and include novel challenges, not just emulate our habitual activities. This ensures that the beneficial effects of cognitive exercise extend to the right hemisphere, which is definitely worth saving. Without overt advice from brain research, some of humanity’s most powerful minds understood this intuitively and lived their lives accordingly. Think of the violin-playing Albert Einstein or of the landscape-painting Winston Churchill, whose recreational activities were distant from their respective “day jobs” of theoretical physics and high-power politics.

Because I do not want to overstate my case, let us look at cognitive exercise from a realistic perspective. Does a life of mental vigor offer a guarantee against age-related mental decline? Unfortunately, no. Just as a healthy physical lifestyle offers no guarantee against a heart attack, influences not under our control (at least not yet), such as genetic heritability, play an important role in brain health. Genetic vulnerability may explain why some of civilization’s greatest minds, such as Isaac Newton and Immanuel Kant, succumbed to the effect of cognitive decline with age, despite their highly cerebral lifelong pursuits.

Clearly, cognitive exertion is not a magic bullet to halt dementia in its tracks. But as a lifestyle ingredient, it appears to slow down cognitive decline, perhaps significantly. Evidence suggests that a vigorous mental life may play a dramatic, if less than definitive, role in helping maintain cognitive well-being at advanced stages of life. This idea is working its way into the mainstream of science. A comprehensive study of successful aging sponsored by the MacArthur Foundation in the 1990s has shown that advanced education—and thus a greater propensity toward mentally demanding activities throughout life—is associated with a sharper mind well into old age. Unsurprisingly, the protective effect of cognitive stimulation in aging is not limited to humans but appears to be present in other species. A study this year showed that learning ability in older dogs—in this case, beagles—is preserved by behavioral enrichment and dietary fortifications.6 As a responsible owner of a three-year-old bullmastiff, I found this report particularly reassuring.

But back to humans. Should we consider regular cognitive exercise as part of our lifestyle? This is still a relatively new concept, not yet fully embraced by health professionals concerned with aging. But it is definitely gaining a toehold: in 2002 no less an authority than The Journal of the American Medical Association published a report showing that a program of cognitive exercise significantly slows mental decline in aged individuals.7

Evidently, it is even possible—up to a point—to have a sound mind in an Alzheimer’s brain.8 This seemingly oxymoronic proposition begins to make sense if one considers the possibility that the pathological changes of dementia, such as Alzheimer’s plaques, and the beneficial effects of cognitive exercise, which stimulates the growth of new neurons, may occur at the same time, act in opposite directions, and cancel each other, at least for awhile. This protective effect may also explain the ongoing productivity seen in some people affected by early dementia, for example, the Basque sculptor Eduardo Chillida or the Dutch-American artist Willem de Kooning.

A Personal Trainer for Your Brain?

All this leads to an intriguing proposition. If vigorous cognitive exertion is so good for you, why not develop scientifically informed cognitive exercises for the brain? As we saw, the effects of cognitive exertion can be specific, benefiting those parts of the brain directly called on by the cognitive challenge at hand. To achieve optimum benefits, participants in training could enroll in an individualized, custom-tailored cognitive workout consisting of a range of exercises that each target a particular aspect of the mind.

Does this sound fanciful and “new age”? Consider that, a half-century ago, the idea of systematic, highly structured physical exercise for the general public (as opposed to sporting elites) sounded fanciful, too, but look where we are today. Health clubs and cardiovascular centers are ubiquitous, attracting a range of users. It took physical fitness several decades to move from a fringe avocation to a mainstream lifestyle activity, but popular culture is evolving faster in the 21st century. I would not be surprised if, within the next few years, cognitive fitness were to proliferate as an organized mainstream activity.

The well-worn adage says, “Nothing is more practical than a good theory.” It is a fitting characterization of the brain research trends we have been analyzing. It would not be so surprising, then, if our budding understanding of the protective role of neuroplasticity in the aging of the brain, coupled with new knowledge of the brain’s hemispheres and their roles, resulted in the emergence of a new mass culture lifestyle trend of cognitive fitness.


  1. Sun, T, Patoine, C, Abu-Khalil, A, et al.“Early Asymmetry of Gene Transcription in Embryonic Human Left and Right Cerebral Cortex.” Science 2005; 5729: 1794-1798.
  2. Herbert, MR, Ziegler, DA, Deutsch, CK, et al. “Brain asymmetries in autism and developmental language disorder: a nested whole-brain analysis.” Brain 2005; 128(1): 213-226.
  3. Goldberg, E, and Costa, L. “Hemisphere differences in the acquisition and use of descriptive systems.” Brain and Language 1981; 12: 144-173.
  4. Goldberg, E, Podell, K, Harner, R, et al. “Cognitive Bias, Functional Cortical Geometry and the Frontal Lobes: Laterality, Sex and Handedness.” Journal of Cognitive Neuroscience 1994; 6: 276-296.
  5. Eriksson, PF, Perfilieva, E, Bjork-Eriksson, T, et al. “Neurogenesis in the adult human hippo-campus.” Nature Medicine 1998; 4: 1313-1317.
  6. Milgram, MN, Head, E, Zicker, SC, et al. “Learning ability in aged beagle dogs is preserved by behavioral enrichment and dietary fortification: a two-year longitudinal study.” Neurobiology of Aging 2005; 26: 77-90.
  7. Ball, K, Berch, DB, Helmers, KF, et al. “Effects of Cognitive Training Interventions With Older Adults.” JAMA 2002; 288: 2271-2281.
  8. Wilson, RS, Mendes de Leon, CF, Barnes, LL, et al. “Participation in cognitively stimulating activities and risk of incident Alzheimer’s Disease.” Journal of the American Medical Association 2002; 287: 742-748.