David Ortiz, the Boston Red Sox’s left-handed batting hero, is one of the American League’s leaders in slugging percentage (.604). But Ortiz is not in the lineup when the New York Yankees’ lefty  flamethrower, Randy Johnson, is the starting pitcher. Ortiz simply cannot hit those sizzling left-handed fast balls for which Johnson, a five-time Cy Young Award winner, is famous. Baseball lore has it that left-handed batters have more trouble hitting off left-handed pitchers than right-handed batters do against right-handed pitchers. 

Whether or not that particular belief is well founded, speculation about what advantages and disadvantages may be conferred on the approximately 10 percent of Americans who are “not-right-handed” has long fascinated scientists.* Knowing that each side (hemisphere) of the brain controls movements of the opposite side of the body, some scientists have looked to handedness to yield clues to how the hemispheres function. Adding greatly to the interest are reports that three times as many left-handed people have schizophrenia, bipolar disease (manic-depressive illness), or autism as would be expected from the prevalence of left-handedness in the population at large. A disproportionately high prevalence of left-handedness has been found, as well, in people with dyslexia and stuttering, in people with math and music prowess, and among people in intellectually demanding professions. But, not surprisingly, some of the proposed linkages have been challenged by other studies that use different designs, so scientists are left without definitive answers. 

Much of the scientific exploration of handedness and hemisphere “dominance” was initially inspired by the work of Norman Geschwind, M.D., and Albert Galaburda, M.D., at Harvard University. In 1985, the two men published a comprehensive article postulating connections between hemisphere dominance and certain disabilities and abilities, posing hypotheses based on clinical observations and known anatomical and physiological data. The interest ignited by the article has led to decades of further studies. 1

Hemisphere “dominance” does not mean that one side of the brain controls the other side. Rather, a person’s dominant hemisphere is the one in which the brain’s language processes and the motor capacities that facilitate speech reside. As emphasized by Michael Gazzaniga, Ph.D., a former Dartmouth College researcher now at the University of California, Santa Barbara, who has studied functions and interactions of the hemispheres, for most people the left hemisphere is dominant for language, including speech, and the right is specialized for tasks such as recognizing faces, grasping spatial relations, and paying attention. A thick bundle of nerve fibers, called the corpus callosum, connects the hemispheres of the brain, providing us with the sense that our brain is integrated. But do right-hemisphere skills explain why left-handed baseball, tennis, and handball players have been found to display better coordination between vision and spatial-motor tasks? 

Geschwind and Galaburda noted an unexpectedly high prevalence of non-righthandedness in people with certain abilities and disabilities, and related these observations to what was known about hemisphere dominance. For instance, they reported that non-right-handedness seemed to be more common in certain groups such as architects and mathematically gifted children. They also noted a higher frequency of both left-handedness and disorders such as autism and learning disabilities among males. Because studies by others indicated that the brain develops later in males than in females, and that the left hemisphere matures later than the right, investigators have speculated that males might be at increased risk of developmental disorders with as-yet-unknown causes occurring at the time of left-hemisphere development. In addition, according to Geschwind and Galaburda, immune diseases (particularly autoimmune bowel disease and juvenile diabetes) occurred almost three times more frequently in non-right-handed people than in the population in general. They emphasized that their findings did not imply that non-right-handedness caused either the talents or the disorders. Instead, they suggested that non-right-handedness was a marker—a sign or indication—of an alteration in the brain’s normal asymmetry of the hemispheres. Abnormalities in brain asymmetry, they suggested, might result from disordered migration of brain cells. 

Some attempts to relate handedness to hemisphere dominance have fostered serious misconceptions. For instance, a popular but wrong notion is that all left-handed people are right-hemisphere dominant, because that hemisphere controls the movement of the body’s left side. This notion mistakenly assumes that dominance for language is inextricably linked to movement control (and so must be in the same brain hemisphere). Research has exploded the myth that left-handed people are invariably right-hemisphere dominant. Instead, like 97 percent of right-handed people, 70 percent of non-right-handed people are left-hemisphere dominant. 

But questions remain. For instance, what are the implications, if any, for the 3 percent of right-handed people and 30 percent of non-right-handed people for whom language function either resides in the right hemisphere or is divided between the hemispheres? For that matter, why and how does the brain usually develop asymmetrically, with language function most often residing in the left hemisphere? What are the implications, if any, for those people who have more symmetrical hemispheres? These complex questions have baffled neuroscientists, but now a plausible and intriguing attempt at an explanation comes from the field of genetics. 


Geneticist Amar Klar, Ph.D., at the National Cancer Institute, developed his theory based on yeast research begun at the Cold Spring Harbor Laboratory on Long Island. He was recruited there by its president, Nobel laureate James D. Watson, Ph.D., as an investigator with an ability to think “outside the box.” Klar was intrigued with asymmetries in organ placement and in the human brain’s hemispheres. Moreover, he was perplexed by the failure of the two prevailing theories, independently advanced by two psychology researchers more than 25 years earlier, to explain fully the inheritance of hand and hemisphere dominance. Both earlier theories proposed that a single gene directed both right-handedness and left-hemisphere dominance. 

How did these theories explain the occurrence of left-hemisphere dominance in most people, but non-right-handedness in 10 percent of the population? Marian Annett, Sc.D., a psychologist at the University of Leicester in the United Kingdom, put forward the “Right Shift” theory in 1972. The presence of a “Right Shift” (RS) gene biases speech toward the left hemisphere and incidentally weighs handedness toward the right, she hypothesized. Chance factors, however, influence both of these independently, resulting in a continuum for both handedness and hemisphere dominance. The RS gene, she further suggested, could impair the speech-related cortex of the right hemisphere. Moreover, a mutated RS gene could produce an anomaly in cerebral dominance that was hypothesized by other researchers to occur in people with schizophrenia and autism. The mutated RS gene might cause a loss of directional coding, impairing either hemisphere at random. When a person inherited a mutated RS gene from both parents, both brain hemispheres would be impaired 50 percent of the time. This theory is consistent with the estimates of schizophrenia in relatives.2 A person with a parent or sibling with schizophrenia has about a ten percent risk of developing the disease, compared to a one percent risk for someone without an affected family member. 

I.C. McManus, a psychology researcher at University College, London, advanced a related hypothesis. In 1985, he suggested the existence of two forms of a gene for brain lateralization, a “dextral” (right) and a “chance” form. People inheriting the chance form of the gene had a 50-50 (that is, random) chance of being either right-handed or non-right-handed. According to McManus, the principal difference between his theory and Annett’s lay in her concept of a continuum versus his notion of two discrete categories. But both theories invoke a single gene, a random component, and classic Mendelian genetic inheritance.3

Although both theories offered an explanation of hand and hemisphere dominance, and even attempted to explain data that showed a higher prevalence of certain brain diseases in non-right-handed people, the theories failed to fully explain handedness and hemisphere asymmetry. Still other researchers posited that non-right-handedness might be the result of interactions among multiple genes. Curiously, in this debate over a genetic theory, geneticists for the most part sat on the sidelines. But Klar, whose specialty was yeast genetics, made some observations that he thought might help to explain hand and hemisphere dominance in humans. When Klar described his theory to human geneticists, the presentation generated excitement—and many questions. 


“Just as our internal organs are usually asymmetrical, with the heart on the left and liver on the right,” Klar explained, “the two sides of the brain also are usually asymmetrical.” How does this happen? The answer lies within our DNA. “Everyone thought DNA strands were passed on randomly to daughter cells during mitosis (cell division),” said Klar, but if that process turned out not to be random at some critical stage, it could provide the opportunity for daughter cells to differentiate, to become different from one another in some key way. This is the crux of Klar’s theory, and, if he is right, it could become a fundamental tenet of developmental biology.



Klar hypothesizes that when cells are dividing during development of the embryo, one daughter cell receives both copies of chromosome 11 that carry the original “Watson” DNA strand, intertwined with the copy of the “Crick” DNA strand. The other daughter cell receives both copies of chromosome 11 that carry the original Crick DNA, intertwined with the copy of the Watson DNA strand. The original Watson DNA strand on chromosome 11 is inherently different from the original Crick strand because it contains an activated “hemisphere dominance” gene, whereas in the original Crick DNA strand this gene for hemisphere dominance is silenced. © 2005 Chistopher Wikoff Courtesy of Amar Klar

Klar explained that the basis for his theory came from his work on inheritance patterns in a type of yeast. When the yeast cell divided, the two resulting cells spontaneously alternated between male and female sexual types. But how did the cells change or differentiate in this way if the DNA, the genes passed on to both cells, was the same? Klar’s answer challenges two fundamental genetic assumptions. 

Assumption one: Daughter chromosomes are always identical to each other, as implied by the work of Watson and his 1962 Nobel Prize co-awardee, Francis Crick. (Klar says that sometimes the two daughter chromosomes differ in the way they express genes.) Assumption two: The two DNA strands are passed on randomly to the two cells that result when a cell divides during embryonic development. (Klar maintains that distribution of DNA strands is not always random.) 


Klar points to a diagram of the pattern of DNA-strand distribution that, he believes, may explain the usual development of asymmetrical brain hemispheres. © 2005 Chistopher Wikoff Courtesy of Amar Klar

Klar’s theory is difficult even for geneticists, so what follows is a greatly simplified but still challenging version. Begin with some basics. An embryo grows as each cell undergoes mitosis, dividing to form two daughter cells. Each of the daughter cells has a full set of DNA, distributed in 23 pairs of rod-like structures called chromosomes in the nucleus of every cell. Each chromosome derives half of its genes from the mother’s egg and half from the father’s sperm. There are about 30,000 human genes distributed among the chromosomes. 

Klar explained: “Watson and Crick noted at the time they described DNA’s structure that the specific pairing of the two DNA strands suggests how the genetic material can be copied.” When a chromosome is replicating itself, its two strands of DNA unwind. After unwinding, each strand of DNA is used as a template to synthesize a new complementary DNA strand. 

Klar assigned to one DNA strand the name “Watson” and to the other the name “Crick.” When the two copies of each chromosome are made, one duplicated chromosome contains the original Watson DNA strand plus a newly synthesized Crick strand, and the other duplicated chromosome contains just the opposite, an original Crick DNA strand intertwined with a new Watson strand. This distinction is critical to understanding Klar’s theory. 


Geneticists have long assumed that during mitosis the duplicated chromosomes are randomly distributed to each new daughter cell. But Klar hypothesized that, at least for chromosome 11, which appears to be involved with brain formation in the embryo, the distribution is not random—at least not at the pivotal point in development when the brain hemispheres are being formed. Instead, according to Klar, one daughter cell receives both copies of chromosome 11 that carry the original Watson DNA strand, intertwined with the copy of the Crick DNA strand. The other daughter cell receives both copies of chromosome 11 that carry the original Crick DNA, intertwined with the copy of the Watson DNA strand. 

Why is this event critical for determining which brain hemisphere will be dominant for language? According to Klar’s theory, the original Watson DNA strand on chromosome 11 is inherently different from the original Crick strand, not just the usual upside-down copy. The original Watson DNA strand contains a “hemisphere dominance” gene that is activated, whereas in the original Crick DNA strand this gene for hemisphere dominance is silenced. Such silencing occurs with other genes, but it is not necessarily strand-specific. In what Klar called a “patterned” distribution, both copies of chromosome 11 that contain the original Watson DNA strand, with its activated gene for hemisphere dominance, always go to the daughter cell on the left. As a result, the left hemisphere becomes dominant for language. The daughter cell on the right, which receives both copies of chromosome 11 that contain the original Crick DNA strand, is silent with regard to hemisphere dominance. 

How do both copies of chromosome 11 that contain the active gene for hemisphere dominance get selectively delivered to the daughter cell on the left, while both copies of chromosome 11 that contain the silenced hemisphere dominance gene get delivered to the cell on the right? Klar hypothesizes the existence of a second gene, which he named the RGHT gene. “This gene acts like a pair of tweezers,” Klar explained. “It recognizes a site on chromosome 11 that has the original Watson DNA strand containing the activated dominant hemisphere gene. I call this site a ‘SEG’ (for segregation) site.” The RGHT gene docks with the SEG site and pulls both copies of chromosome 11 that contain the original Watson DNA, with the activated hemisphere dominance gene, to the daughter cell located on the left side of the embryo.4

Do all chromosomes get distributed to daughter cells in this selective way? Klar thinks not. Some or most chromosomes may be distributed randomly. Evidence for simultaneous patterned and random distribution of chromosomes to daughter cells was recently confirmed in studies of mice. This discovery led Klar to revisit results of a mouse study that suggested that chromosome 7 was distributed in a patterned, rather than random fashion. Klar’s excitement soared when he realized that more than a third of the DNA sequences in mouse chromosome 7 were similar to those in human chromosome 11, the chromosome that Klar hypothesizes directs left hemisphere dominance in humans. 

When does the patterned segregation of DNA strands to daughter cells, crucial to Klar’s theory, actually occur? Probably not as early as the first cell division of the fertilized egg, said Klar, because, if it did, the body’s organs would be randomly distributed, and this is not the case. Instead, “it probably occurs when the three layers of cells in the embryo become differentiated.” The brain, skin, and peripheral nervous system cells of the limbs and torso arise from the ectoderm, the outermost layer of the embryo. The body’s connective tissue, muscle, blood, and bones, develop from the mesoderm (middle layer), and the body’s organs and intestinal tract develop from the endoderm (innermost layer). 

If Klar’s theory is proved correct, it could become a new model for developmental biology, explaining how cells differentiate when they divide and how asymmetric cell division occurs throughout the body. According to Klar: “Asymmetric cell division sometime during embryonic growth usually leads to the development of a dominant left-brain hemisphere, where language is processed. If hemisphere dominance was like Yogi Berra’s adage, ‘if you find a fork in the road, take it,’ you would expect a random distribution—50 percent of the time the right hemisphere would be dominant for language—and that is not the case.” 

Then what occurs differently in the 3 percent of right-handed people and 30 percent of non-right-handed people who have language function in the right rather than the left hemisphere, or more equally distributed to both hemispheres? Klar hypothesizes that the RGHT gene occurs in two forms, which is key. Similar to many other genes, one form is dominant (R) and the other recessive (r). In contrast to the dominant form, which pulls both copies of chromosome 11 with the original Watson DNA to the daughter cell on the left, the recessive form “lacks a signal for choosing direction,” he said. About 84 percent of the population has the dominant form of the RGHT gene, inheriting this form from both or at least one parent (in genetic terminology, RR or Rr). This entire 84 percent of the population is left-brain dominant and right-hand dominant, according to Klar’s theory. 

“The left hemisphere controls the right side of the body, and people with left-hemisphere dominance develop a preference for using their right hand for complex tasks,” Klar explained. “The RGHT gene makes a person right-handed by guiding the developmental choice to move in a specific direction.” 


That leaves us with the remaining 16 percent of people who inherit the recessive form of the RGHT gene (rr) from both parents. Because these people lack the genetic signal for choosing direction, Klar explained, “each person has a random (50 percent) chance of being either right- or non-right-handed, and, independently, an equal chance of having hemispheric dominance for language on either the left or right side.” 

Klar contends that the best approach to evaluating his theory is to see, first, whether the theory as a whole—the big picture— holds true. Only then, he maintains, should one worry about the many unanswered details of the model. Whereas conventional molecular geneticists work the other way around, he stresses that for him: “Details come later.”

To test his model of the genetics of handedness, Klar identified families in which both great-grandparents were left-handed and studied the hand preference of the two succeeding generations. “This was the only study, among hundreds, which looked at three generations,” Klar said. He found that when two non-right-handed parents produced right-handed children, the children then produced a similar percentage of non-right-handed children as did their conventional non-right-handed peers. Klar’s model fit a 1940 observation that 50 percent of children of a non-right-handed mother and father are non-right-handed, whereas only 8 percent of children born to a right-handed mother and father are non-right-handed. Klar’s finding argues against handedness being, in part, a learned behavior or the result of stress damage during birth.

“What really excited us,” he continued, “was that the model could explain why 18 percent of identical twins were discordant for handedness”—in other words, why identical twins do not always show the same hand preference. In this situation, both identical twins inherited the rr gene, and each twin had a 50-50—random—chance of becoming right- or non-right-handed. In identical twins and in all others inheriting the rr form of the gene, the theory also explains how the rr form separately and randomly controls the determination of handedness and hemisphere dominance. 


For further evidence to rule out the possibility that handedness was determined by interactions of multiple genes, or behavioral or other environmental influences, Klar decided to observe an environment-free trait: the direction of hair whorls that develop on the top of the head. Klar wanted to know whether hair whorls, like handedness and hemisphere dominance, might be guided clockwise by the RGHT gene and be equally divided between clockwise and counterclockwise in people with the gene’s recessive form. 

Perched at the top of escalators in shopping malls, he observed the heads of riders. He found that 92 percent of the riders whose hair whorls he could see had clockwise hair whorls, fitting like a glove with the prevalence of right-handedness in the population. The percentage of clockwise whorls also fit with a 1927 German study of hair-whorl rotation in families. Next, through more conventional studies, Klar found that people with counterclockwise whorls were equally divided between right- and non-right-handed. So hair-whorl direction was coupled with handedness in right-handers and decoupled (occurred randomly) in non-right-handers. “This establishes that handedness and hair whorl develop from a common” genetic process, and “behavior is not a factor.5 Hand preference is purely genetic,” Klar stated. Now, he says, “a direct test of the association of hair-whorl orientation and brain laterality needs to be done.” 


Hair whorls may be guided to take a clockwise instead of counterclockwise direction by a dominant RGHT gene. When the gene’s form is recessive, however, direction may be equally divided between clockwise and counterclockwise. Klar believes that hair-whorl direction is coupled with handedness in right-handers and decoupled in non-right-handers. There also appears to be an association between hair-whorl direction and male homosexual preference, so there may be a common genetic factor. © 2005 Chistopher Wikoff Courtesy of Amar Klar 

Working on his own time while vacationing at the beach, Klar then stumbled on an association between hair-whorl direction and male homosexual preference. He realized that in the area where he was “most of the people were males showing stereotypical homosexual behavior.” Klar noted the direction of the men’s hair whorls and found them to be counterclockwise in almost a third of the men. This contrasted with the eight percent prevalence of counterclockwise whorls in the general population of males. During two additional visits to the same beach during the year, he found a 3.6-fold excess prevalence of counterclockwise hair whorls among groups with a preponderance of homosexual men compared to the rate in populations of men in general that he had studied elsewhere. From this, Klar concluded that, in a significant proportion of homosexual men, sexual preference might be influenced by a genetic factor that also controls hair-whorl direction. 


Klar believes his theory also may explain why there is a threefold higher prevalence of non-right-handedness in people with schizophrenia and people with bipolar disorder. Some studies of families with several members who have either schizophrenia or bipolar disease suggest abnormalities related to hemisphere asymmetry. In particular, the affected family members with psychoses seem to have brain hemispheres that are more symmetrical in terms of their functioning, with language processes present in both hemispheres. While many studies point to some genetic basis for psychosis, no gene mutation has been identified, despite decades of research. 

Klar’s survey of this literature focused on studies that might associate chromosomal rearrangements with psychosis. In one comprehensive study of multiple generations of a large Scottish family, a “translocation” had occurred in chromosome 11. That is, chromosome 11 breaks in two, and each part recombines with a part of chromosome 1. Such hybrid chromosomes have persisted across more than five generations of the family. Nine family members were diagnosed with schizophrenia and nine with bipolar disorder, but none of another 18 members of the family with this chromosomal translocation has developed either disease. 

Klar speculated that in the family members with the translocation, the SEG site might become located on chromosome 1 instead of 11. As a result, the RGHT gene cannot perform its function of guiding the distribution of the original Watson DNA strand, with the activated hemisphere dominance gene, to the daughter cell on the left side of the embryo. Instead, distribution is random. This results in one-half of the family members with the translocation developing left-hemisphere dominance, while the other half develop language function more symmetrically in both hemispheres. In this other half, consequently, according to Klar’s theory, psychosis is related to hemisphere symmetry. 

“Importantly,” Klar noted, “this model suggests that schizophrenia and bipolar disease in these families do not develop because of a mutation in any gene, or to small mutations in several interactive genes.” Instead, the translocations disrupt the patterned segregation of DNA stands of chromosome 11 chains, which disrupts development of normally asymmetrical hemispheres. Klar’s theory does not attempt to explain how hemisphere abnormalities cause schizophrenia or bipolar disease. He leaves that key aspect to neuroscientists. Many studies using fMRI, however, have shown that patients with psychoses have less hemisphere asymmetry than people without psychoses. Klar found several additional studies that described other chromosomal translocations that were partially correlated with psychosis, and chromosome 11 was the only one common to all. “This blew me away,” he said, adding that the translocation of chromosomes 11 and 1 provides the best evidence, to date, to support a genetic cause for schizophrenia and bipolar disorder. 

Klar emphasized, however, that psychosis in the general population surely does not develop from a translocation of chromosome 11. Rather, he hypothesized, psychosis may result in a small percentage of people who inherit the rr gene and develop more functionally symmetrical hemispheres.6 Ordinarily, the left hemisphere, according to Michael Gazzaniga’s recent thinking, may create a unified sense of self by monitoring and interpreting the behaviors of networks of brain cells throughout the brain. The left hemisphere ordinarily interprets all these inputs and “weaves them into stories to form the ongoing narrative of our self-image and beliefs.”7 Could this function be altered in people with psychosis? 


Although intriguing, Klar’s theory, like others that suggest a genetic basis for hand and brain hemisphere dominance, has postulated the existence of a gene or genes that “ought” to exist, based on what has been observed about inherited characteristics. No such gene has been identified. 

“Watson keeps telling me to find the RGHT gene,” said Klar. “Usually, in genetics, people find a gene associated with a disease or a trait, and then try to figure out what the gene does. At least in this case, we have a good guess about the gene’s function, that it non-randomly distributes (at least) chromosome 11 chains during cell division.” So, for now, Klar’s theory, which began with yeast and then turned to ascertaining how internal organs become located on the left or right side of the body, provides an intriguing new framework for considering how hemisphere and hand asymmetry may develop. The model may lead neuroscientists to pose new questions about the neurobiology of disease, and maybe even homosexual preference. Could Klar’s theory also help to explain the excess prevalence of non-right-handedness among people with autism or people with superior talents, skills, and intellectual abilities? If neuroscientists understand how brain asymmetry develops, or fails to develop, will this insight lead to research approaches in the neurobiology of certain conditions? 

For, even if the genetic basis of hemispheric dominance is determined through this or some other model, Klar reminds us, we then have a long way to go to discover the far-reaching implications of the theory for developmental biology in general and for brain development in particular.   



A list developed in 1940 by David C. Rife is considered the standard for determining hand preference, although it is rarely used in its entirety. Rife stipulated that a person is not purely right or non-right-handed unless he uses the same hand to:

  1. throw a ball
  2. shoot marbles
  3. bowl
  4. hammer
  5. write
  6. saw
  7. sew
  8. use a spoon
  9. cut with a knife
  10. cut with scissors


  1. Geschwind, N, and Galaburda, M. “Cerebral lateralization.” Archives of Neurology 1985; 42: 428-456.
  2. Annett, M. “The theory of an agnostic right shift gene in schizophrenia and autism.” Schizophrenia Research 1999; 39(3): 177-82.
  3. McManus, IC. “The inheritance of left-handedness.” CIBA Foundation Symposium 1991; 162: 251-67.
  4. Klar, AJS. “An epigenetic hypothesis for human brain laterality, handedness, and psychosis development.” Cold Spring Harbor Symposia on Quantative Biology 2004. LXIX: 499-506.
  5. Klar, AJS. “Human handedness and scalp whorl direction develop from a common genetic mechanism.” Genetics 2003; 165: 269-276.
  6. Klar, AJS. “A genetic mechanism implicates chromosome 11 in schizophrenia and bipolar diseases.” Genetics 2004; 167: 1833-1840.
  7.  Gazzaniga, MS. The Ethical Brain. Dana Press. Washington, DC, 2005: 148.


About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Robert Malenka, M.D., Ph.D., Stanford University School of Medicine
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.