Were We Born This

Parenting Children With Asperger's And High-functioning Autism

Autism Therapy

Get Instant Access


My young Labrador retriever, Zoe, has an endearing (if sometimes maddening) need to be near her human family. As I write these words, she is sitting close to me, her muzzle resting on my thigh. It does no good to push her away. She needs to be this close. Zoe is affectionate to a fault. She is also a terrible beggar, willing to eat anything faintly resembling food. She will also play fetch as long as one of her humans can tolerate throwing her slimy stuffed bear across the yard.

Dog fanciers know that breeds have characteristic behaviors. The Great Dane is prized for its courage as well as its size, the tiny Chihuahua is irrepressibly lively, and the golden retriever is adored because it is adoring. Newfoundland puppies would rather swim in a bowl of water than drink it. Border collies, the product of centuries of breeding in Scotland and England, are born herders. The mere sight of sheep elicits a hypnotic stare and a low crouch. Yellow labs become clinically depressed when separated from their humans.

Yet, just a few thousand years ago, there were no dogs as we know them. If evolution is such a slow process, how could there be so many distinct breeds today? About 10,000 years ago, humans probably began to keep the cubs they found after killing wolves. Genetically programmed to belong in a group, the cubs must have accepted membership in the human clan. At first men probably used the descendants of these early canine partners mostly in hunting. As human culture matured and men became farmers, dogs, animals that possess a highly accurate ability to recognize outsiders, became valued as guards. Eventually, we came to value them for their company. In a sense, slaves became servants and, in time, friends.

Centuries before Mendel, dog breeding was already being widely prac ticed in many human cultures. In his great book, The Origin of Species (1869), Darwin cites the success of dog breeders in artificially selecting desired features in their animals to support his thesis that species evolve through natural selection. In the world of dog breeding, man is a surrogate for nature. Exerting a pressure far more intense than nature, breeders select the qualities that they want to perpetuate and permit only the exemplars to breed. Darwin was deeply interested in the success with which breeders could select for behavioral as well as physical traits. In another book, The Descent of Man, he speculated that although breeding had not caused dogs to gain in cunning (a trait he thought was more highly developed in the wolves from which they descended), they must have "progressed in certain moral qualities such as in affection, trustworthiness, temper, and probably in general intelligence." In Zoe's case, Darwin's speculation seems correct.

There is not necessarily a direct correlation between a highly prized physical or behavioral trait and a particular gene. But generations of breeding, often inbreeding of closely related dogs, has through trial and error made breeders quite confident of what they can expect in a litter. The reason for this is that the various breeds share many alleles (variations of particular genes) in common. One can think of breeding as an effort to identify, select for, and propagate slight improvements in phenotype that reflect small changes in the breed's gene pool about which the breeder is, albeit inchoately, knowledgeable.

Dr. Jasper Rine, a geneticist at the University of California at Berkeley, is among a handful of scientists who deserve credit for moving dog breeding from high art to hard science. Using the same techniques that permit the creation of detailed maps of the human genome, he and his colleagues started the arduous process of mapping the dog genome. Since the project began in 1990, they have mapped thousands of genetic markers more or less equally distributed across the 78 dog chromosomes. The fact that dogs have nearly twice as many chromosomes as humans does not mean that they have more genes. The number of chromosomes in each species varies widely among mammals, but the total number of genes is probably roughly the same. On each of the 78 dog chromosomes, the genes are arranged in "linkage groups" that are in many cases highly similar to the order of comparable human genes on our chromosomes.

There are several obvious reasons to study the dog genome. Of the more than 150 modern breeds, many are burdened with genetic disorders that constitute excellent models for the study of comparable human diseases. Dobermans and Scotties are at risk for hemophilia, Bedlington terriers may have abnormalities in copper metabolism akin to a rare disorder in humans called Menke's disease, Labrador retrievers and several other breeds are at high risk for congenital hip dysplasia, and beagles sometimes suffer from genetically caused seizure disorders. These are usually either X-linked (hemophilia) or autosomal recessive disorders (conditions in which the presence of a single allele in the animal is harmless, but in which the inheritance of a mutated allele from each parent causes the disorder in offspring). Unless one has a way to identify animals who carry the disease-causing recessive allele, having pups at relatively high risk for a genetic disorder is a largely unavoidable consequence of extensive inbreeding. Just as is the case with human genetic disorders, once dog geneticists have mapped the alleles that cause illness, it will be straightforward to develop tests to identify animals that carry the various recessive alleles. Breeders will then be able to avoid mating a pair of carriers, which will be a boon both to them and to people who want to purchase the animals. They will eventually be able to certify that a pup has been tested for a group of genetic disorders and has been determined to be risk-free. In 2000 Dr. Elain Ostrander, a geneticist in Seattle who once worked with Dr. Rine and who is now the coordinator of the Dog Genome Project, reported that a group of about 15 laboratories had succeeded in correlating most of the dog genome with corresponding locations in the human genome. Along the way more than 20 canine disease genes have been mapped. Among the most important is the discovery of the gene that causes narcolepsy in Doberman pinschers, a discovery that may help understand the disease in humans.

Although eager to help in the conquest of genetic diseases among dogs, behavior geneticists seek a more elusive prize. They think that the behaviors which characterize many breeds are the product of relatively few genes, and that, once armed with a fairly informative genetic map, they will be able to track down those genes. Reversing the centuries-old strategies of breeders, they want to cross dogs of sharply different breeds that are also characterized by widely different behaviors.

Although all dog breeds are part of the same species, generations of selection through controlled breeding have almost certainly created dra matic differences in the frequency with which certain gene variants are distributed across breeds. Variation among dog breeds is, for example, much greater than is variation across human groups. Take size. Almost all adult human males weigh between 125 and 250 pounds. Men from some ethnic groups, say Northern Europeans, tend on average to be as much as 50% heavier than men from other groups, but the range is not much greater than twofold. In dogs the situation is much different. Irish wolfhounds are 50 times (5000%) heavier than Pekinese. Even so, there are no insurmountable barriers to artificially mating these two breeds and then using genetic mapping techniques to isolate genes that exert great influence on growth and adult size.

To hunt for genes that shape dog behavior, geneticists can mate animals from breeds with two sharply different behaviors, water-loving or herding or loyalty, for example, and then determine which offspring show which behaviors. They can then breed those animals and again look for the trait in offspring. In each instance, the scientists correlate the presence or absence of the phenotype in the animals from the second generation with the presence or absence of genetic markers that they know must bracket the chromosomal segment on which the gene that influences the particular behavior must reside. Over time, they can gradually narrow the stretch of DNA until they define the region that is always associated with the trait. That region will contain the gene.

If scientists succeed in finding genes that define dog behavior, they will immediately trigger two profoundly interesting questions. What are the comparable genes in humans and what are their functions? For example, once a gene associated with the herding instinct in dogs has been cloned, it may take only a matter of minutes to search a database of human genes to determine whether there is a human analog, which there almost certainly will be. Of course, the existence of a comparable gene in humans does not mean that it plays any discernible role in shaping human behavior, but many think that the human analog could influence some similar behavior, an intriguing and, doubtless for some, a discomforting possibility. Imagine if Darwin was correct when he suggested that in the "deep love of a dog for his master" we discern a "distant approach" to religious feelings in humans! Might there be human genes for religiosity? Fortunately, for those of us who squirm at such analogies, it will be many years before we make these kinds of connections.


We still know almost nothing about the role played by particular genes in shaping the nuances of human personality. The main obstacles to exploring the matter are the same as those that have hobbled efforts to discover genes that strongly influence risk for psychiatric disorders. Observers cannot consistently agree on the boundaries of subtle behavioral phenotypes in humans. Where are the lines that separate those who are "unusually" shy from those who are "on the shy side" or the individual who is relatively at ease among strangers from the natural extrovert? Can such boundaries ever be drawn? How are personality traits shaped by climate, cultural and religious upbringing, education, gender bias, racial discrimination, and myriad other forces? Is not a trait like introversion or extroversion heavily influenced by these and a host of other factors? Of course.

Still, we all routinely use and are comfortable with labels that capture the essence of human personality. "She's shy" or "He's the life of the party," we often say. Such phrases can even extend to ethnic groups. We speak of "dour Scots" and "talkative Italians." Deeply embedded in our language, such phrases suggest that human personality has a few distinctive features and that they are heavily influenced by genes.

Good behavioral geneticists are the first to agree that the challenges to understanding the genetic variance in personality are difficult, but thanks to our ever-growing knowledge of the human genome, they no longer view them as insurmountable. Furthermore, they are sure that a deeper understanding of the genetic components of personality and behavior will yield knowledge of great value. To open a discussion about genes and personality, I start with a common behavioral problem of childhood— bed-wetting—that has long been thought to be correlated with certain environmental stresses, but about which there has recently emerged convincing evidence that it is strongly influenced by genes.

Millions of children (considerably more boys than girls) persistently wet the bed after age seven. For much of this century, primary enuresis, as the child psychiatrists call it, was thought to be due to emotional problems, in many instances serious parent-child conflicts. Many psychiatrists thought that the problem arose out of conflict between a domineering parent and a passive-aggressive child. Families spent long hours with pediatricians and psychiatrists, and both parents and children felt acute em barrassment about bed-wetting, often creating tensions that harmed the daytime routine.

The first studies to suggest that bed-wetting might be genetically influenced appeared about 20 years ago. From responses to questionnaires administered to the parents and grandparents of children who were chronic bed-wetters, it became clear that in many instances the problem was highly familial. One large study showed that in families in which one parent had a problem in childhood with bed-wetting, nearly half his or her children had the problem; when both parents had been bed-wetters, about three-fourths of the children were affected. These results were compatible with the effects of a highly penetrant dominant gene.

In 1995 a group of Danish scientists uncovered powerful evidence that a variation in a gene on the long arm of chromosome 13 caused a substantial portion of serious bed-wetting. In 1973 some of their farsighted colleagues had collected and frozen white blood cells from members of 832 "normal" families, who were special only because each included at least four children, a fact that greatly helps genetic analysis. Twenty years later the researchers who sought to do gene mapping studies in families burdened by primary enuresis sent questionnaires to 655 of these families, asking whether bed-wetting was a problem. Eleven families who acknowledged a strong family history agreed to participate. The researchers compared the DNA markers of the parents who had been bed-wetters to the markers of their affected and unaffected children. All but one of the 23 parents who reported having been bed-wetters transmitted the copy of their chromosome 13 with suspect markers to the affected children, and they passed on the other copy of 13 to their unaffected children. This strongly suggests the action of a dominant gene.

How could a change in a single gene cause a problem like bed-wetting? One of several possibilities is that it could alter the function of a protein that the brain uses to tell the kidney to produce less urine at night. Children who are not as able to concentrate their urine will be unable to avoid urinating for an 8-10-hour stretch. Because kids sleep more soundly than do adults, they will be less likely to awake to avoid wetting the bed.

The genetic hypothesis has greatly changed how parents and health care professionals view bed-wetting. It has been reconceptualized as a physiological disorder that is often caused by inadequate levels of antidi-

uretic hormone. Child psychiatrists now focus on reassuring children and parents that bed-wetting is not a form of acting out, a sign of an anxious temperament, or the result of a personality conflict. Rather than looking for evidence of familial dysfunction, they help families cope with the disorder to reduce the risk that they will become dysfunctional. The discovery of a genetic basis for bed-wetting is impressive, but it falls far short of convincing anyone that human personality is deeply shaped by discernible genetic factors.


What fundamental question might we ask to investigate the role of genes in human personality, and how might we pursue it? Of the many ways in which we can define our own personalities and those of our relatives and friends, none is more basic than our sense about an individual's baseline state of happiness or sadness. Everyone knows people who seem upbeat or happy; most of use know others who strike us as generally downbeat or sad. Could something so fundamental as a predisposition to happiness or sadness be genetically driven? Is happiness heritable? Behavioral geneticists are already probing this question, and a few are already arguing forcefully that the answer is yes.

Among the most interesting and controversial findings are those of David Lykken and Auke Tellegen, psychologists at the University of Minnesota. In studying the heritability of mood, they pursued the time-honored technique of comparing identical with fraternal twin pairs. Their subjects were 1380 pairs of twins born in Minnesota between 1936 and 1955. They assessed "happiness" by using the Well Being scale of the Multidimensional Personality Questionnaire, a self-report inventory that asks respondents to indicate whether they agree or disagree with statements such as "I am just naturally cheerful." They found that identical twins were far more likely than fraternal twins to provide similar answers to similar questions. The cross-twin concordance for identical twins was 0.44, while for fraternal twins it was only 0.08. That is, adult identical twins were far more likely than fraternal twins to record similar answers, a difference strongly suggesting (but by no means proving) that genes have a major influence in shaping one's baseline mood.

Lykken and Tellegen followed the twins for five to ten years and then asked some of them to complete the Well Being scale a second time. This allowed them to compare the responses of each twin at two points in time and, in addition, to compare the responses of one twin at one time to the co-twin at another time. The cross-twin cross-time results among identical twins was about 0.4, much higher than the score of 0.07 recorded by fraternal twins on the same comparison. In an editorial assessing the research, Dean Hamer of the National Institutes of Health, a scientist who does similar work, suggested that the data indicated that the broad heritability of a sense of well-being is 40 to 50%. The implications of attributing such a large fraction of personality to genes are immense. It suggests, for example, that parenting style, economic status, and education have relatively little impact on the child's sense of well-being. Hamer opined that a strong genetic influence on well-being could explain other research indicating that positive life events and socioeconomic success have little impact on subjective feelings about mood. For example, lottery winners are not much happier after winning the jackpot, and people paralyzed by spinal cord injuries are (in the long run) not much sadder than average. Perhaps the most provocative aspect of the research was its suggestion that the best predictor of whether one will be happy 10 years hence is how one feels today, in essence an argument that future happiness is genetically influenced!

We have not found any happiness genes yet, but we may. In the meantime, if happiness is largely a matter of genes, why do we strive so hard to feel better than we naturally do? Lykken and Tellegen have speculated that, "It may be that trying to be happier is as futile as trying to be taller and therefore is counterproductive." A more hopeful answer is probably that we all have a range of moods, and that it might be possible for one to work to maximize the happiest possible frame of mind.

The Minnesota happiness study is only a beginning, a statistical observation based on analyzing a pile of questionnaires. Is there any evidence of particular genes that can significantly affect personality? Doing linkage studies on inbred strains of mice, some scientists have described a pheno-type called "emotionality" and have traced its expression to the effects of just three genes. No such work has been done in humans, but two scientific papers have made (as yet unverified) claims that variants in a single gene can determine aspects of personality such as "novelty-seeking" and the ease with which one makes friends.


The work on the genetics of novelty-seeking also began with a questionnaire. Robert Cloninger, a leading behavioral geneticist at Washington University, created the tridimensional personality questionnaire (TPQ), which is intended to assess four broad categories of temperament that he calls novelty-seeking, harm-avoidance, reward-dependence, and persistence. He did this to pursue his theory that differences in temperament among people are strongly influenced by variations in the production or processing of a neurotransmitter (a chemical messenger that sends signals from one brain cell to another) called dopamine.

In 1996 some Israeli scientists reported that among 124 unrelated individuals, those subjects whose responses to the TPQ placed them high on the novelty-seeking scale were also much more likely than expected to have a particular version of the dopamine D4 receptor gene. The finding was in keeping with other knowledge about dopamine: (1) In animal studies, dopamine levels have been shown to affect exploratory behavior; (2) variations in dopamine metabolism affect how people respond to cocaine; (3) humans vary widely in how their brain cells take up the neurotransmitter; (4) the dopamine D4 receptor is found on a more restricted set of brain cells than other dopamine receptors and may be the site of action of a drug called clozapine. The authors posited that a gene variant that drove novelty-seeking might be more efficient in bringing dopamine into brain cells than is the common version.

The report on novelty-seeking was all the more impressive because it was accompanied by findings from another group that, working independently of the first, had found essentially the same results. Using a different questionnaire, the second team correlated the responses of 315 people with whether or not they had the version of the D4 receptor that the first research group had associated with novelty-seeking. The "extroversion" scores were significantly higher among those who had the same gene variant that the other researchers had associated with novelty-seeking. This was especially interesting because the second study population was quite different from the first. Almost all (95%) of the group was male and about half were gay. This is because the research was done in a laboratory that is studying the genetics of homosexuality (see Chapter 12), and the blood samples and questionnaire results were readily available. The results could be confounded if gay males turned out to show more novelty-seeking behavior than other people. The studies did not make grand claims. The data suggested that the D4 receptor accounts for about 10% of the genetic variation in novelty-seeking. Yet, it is a breakthrough of sorts. Together, these papers (however modest) constitute the first replicated association in humans of a specific genetic locus with a personality trait.


The first suggestion that there is a gene that directly influences one's sociability grew out of a study of Turner syndrome, a condition affecting about 1 in 2500 girls that is caused by being born with one (instead of the normal two) X chromosome. No boys are born without an X chromosome because for them this is fatal early in development. The lack of a second X is often lethal to the female fetus. No one understands why many female fetuses with Turner syndrome die while many others are born and remain healthy.

Girls with Turner syndrome can have a variety of abnormalities, the most obvious of which is extra skin on the sides of the neck (a consequence of abnormalities in the development of the lymph system) that give it a webbed look. However, most affected infants look normal and are often not diagnosed until much later. The girls do grow slowly, and one of the first hints that something is wrong is that they are unusually short. In the not-so-distant past, one leading human genetics textbook quaintly characterized them as "dainty little girls with pleasant personalities who are well behaved and industrious." Unless they are treated with a drug called oxandrolone or genetically engineered human growth hormone, girls with Turner syndrome usually reach no more than 4 feet, 10 inches in height. The diagnosis is usually made when the girls reach their teenage years and do not menstruate or develop breasts. They have only rudimentary or "streak" ovaries.

Contrary to the textbook description, many girls with Turner syndrome have problems in school. Although a few have completed college and taken on challenging jobs, many make it only partway through a standard course of public education. In one European survey of 126 women with Turner syndrome, only 12 had completed high school, and 21 reported having been schooled in classes for children with special needs.

About one-third of the girls score in the mildly retarded range on IQ tests. Scientists have long been fascinated by the fact that there is a sharp discrepancy in various sub scores of that test. Persons with Turner syndrome tend to be unusually weak in math and to have a very poor sense of direction. They often report that they cannot read maps and have trouble finding items in their homes. As they grow up, some persons with Turner syndrome also have great trouble in social relationships.

Because they have only one X chromosome, girls with Turner syndrome represent an intriguing experiment of nature that scientists have used to explore a still poorly understood phenomenon called imprinting. There is good evidence that the action of particular genes in an individual sometimes differs sharply depending on which parent provided the chromosome. Depending on whether they are transmitted through egg or sperm, some genes may never even be activated. Through DNA marker studies, we know that about 70% of the girls with Turner syndrome have inherited their X chromosome from their mothers and about 30% from their fathers. (This is as expected, because females have two X chromosomes and males have only one.) This provides researchers with a way to ask whether genes on the X chromosome show evidence of imprinting. Do they act differently depending on parent of origin?

Aware of the wide range in social skills among girls and women with Turner syndrome, Dr. David Skuse, a behavioral scientist at the Institute of Child Health in London, and several colleagues undertook a project to determine whether affected girls could be divided into distinct groups in which their level of sociability could be correlated with whether they had inherited a maternal or a paternal X chromosome. After obtaining approval from local ethics committees, the group started their work by identifying persons with Turner syndrome from a national registry in England and sending questionnaires to them. Parents, school teachers, and, if they were 11 or older, the girls themselves, were each asked to respond to different survey instruments. Each respondent was asked to evaluate or self-evaluate how the particular person with Turner syndrome was doing in school and in the more general category of social adjustment. Analysis of the responses from parents (who were surveyed twice about two years apart), teachers, and the girls allowed the researchers to obtain differing perspectives on the same set of issues. The team also performed DNA studies to determine from which parent the sole X chromosome derived.

On every social measure the group of 25 girls whose X chromosome derived from their fathers scored much better than did the group of 55 girls whose X chromosome came from their mothers. Their parents, their teachers, and the 25 girls themselves all gave responses indicating that they were well adjusted. The girls with paternal X chromosomes also had better verbal skills and demonstrated much superior "higher-order executive function skills," measures that may reflect how well one will get on in everyday interactions. Only 4 of the 25 girls whose X chromosome came from their fathers needed to be in special education classes, while 22 of the 55 girls with the maternal X were in special education programs (statistically, a difference of great significance). Also intriguing was that 21 of the 29 girls over age 11 with the maternal X had significant behavioral problems, whereas only 4 of 14 similarly aged girls with the paternal X had such difficulties.

The researchers also devised a social cognition questionnaire that they administered to normal boys and girls as well as those with Turner syndrome. On this scale (in which a higher score indicates more evidence of social adjustment problems), the Turner girls with the maternal X had much higher scores than did those with the paternal X, and normal boys had much higher scores than did normal girls. The boys scored about the same as did the Turner girls with the paternal X.

From these findings the researchers concluded that they had uncovered the first solid evidence for the existence of an imprinted gene on the human X chromosome, and that it exerts great influence on the development of social skills. They argue that the gene is only expressed if it comes from the father. Put another way, it will be inactive if it is transmitted through the mother. Because normal girls get an X chromosome from each parent, the genes at this locus will be active. Boys, however, only inherit a single X and always from their mothers. Thus, they might routinely inherit a gene that may have an important role in social cognition, but which never is working to their benefit. In the (somewhat overstated) words of the researchers, this could "explain why males are markedly more vulnerable than females to pervasive developmental disorders affecting social adjustment and language, such as autism." Of interest, the investigators did find three Turner girls who carried the diagnosis of autism. All three had inherited the maternal X! In interviews with the press, Dr. Skuse described the girls with a maternal X as not good at "reading" body lan guage, tone of voice, and other subtle messages. He also suggested that normal girls are "hard-wired" to pick up social skills almost instinctively, while normal boys have to work to acquire those skills.

Science journalists hailed the report as evidence that there is a gene that makes girls more adept than boys in learning social cues. But it is a giant leap to use the results of studies that combine profiles of human personalities with genetic analysis to posit that a single gene explains why boys are at higher risk for autism or have more social adjustment problems than do girls. To those who are troubled by the impact of genetic information on society, the determinism implicit in such speculative leaps seems irresponsible and scary. We will ultimately understand how genes interact with the environment to shape personality. But, our genes are, quite literally, only what we start with. Much good will come of a deeper understanding of the role of genes in the development of personality, but society will not benefit from this knowledge if we overstate its value or tolerate its misuse.

Was this article helpful?

0 0
Aspergers Answers Revealed

Aspergers Answers Revealed

Learn How to Help, Understand amp Cope with your Aspergers Child from a UK Chartered Educational Psychologist. Before beginning any practice relating to Aspergers it is highly recommended that you first obtain the consent and advice of a qualified health,education or social care professional.

Get My Free Ebook

Post a comment