Stranger in the Nest: Do Parents Really Shape their Child's Personality, Intelligence, or Character
A Review Article by Edward M. Miller, Research Professor of Economics and Finance University of New Orleans, New Orleans, La. 70148 emmef@uno.edu (E-Mail), January 17, 1999
How often do parents, having tried to raise the perfect child, one that reflect their values and goals, find that they have a "Stranger in the Nest", a child that turns out not to be one that reflects their values and desires? This happen often, and explain the title of the new book by University of Texas psychologist, David B. Cohen. The basic message of the book is given by the sub-title "Do Parents Really Shape their Child's Personality, Intelligence, or Character", with the answer a resounding no. Instead the book emphasizes the role of genetic and pre-natal events in shaping intelligence and personality.
The book puts the emphasis on the good news for parents whose children have turned out other than as they would wish. It points out how diseases such as schizophrenia, manic-depression, and autism were once blamed on poor parenting, but now they are regarded as due to biological causes that parents can not be blamed for. The evidence for these conclusions comes from family studies, especially those of twins and adopted children. For instance, if one twin is schizophrenic, the other twin is schizophrenic about half of the time, a rate about 50% times as great as the about 1% chance of any particular person being schizophrenic. Likewise, the probability of an adopted child growing up to be schizophrenic is much greater if the birth mother was schizophrenic than is she was not. This evidence points to genetic factors being very important. Similar evidence now exists for conditions such as manic-depression and autism, conditions once blamed on bad parenting.
In trying to make the case for parents and other environmental influences having relatively little influence, Cohen points to cases where a person turned out good in spite of appalling environmental conditions. The black scientist, George Washington Carver is one example. Literature is filled with other examples where people somehow found the strength to overcome obstacles. When one asks the question of where the strength came from, there is often no obvious environmental cause. The very plausible answer is that these individuals were born with it, it was in their genes.
In the other direction, the numerous cases where well-intentioned and good parents produce criminal offspring are described. If outstanding individuals produce poor offspring, and poor parents often produce excellent offspring, differences in parenting within the normal range are unlikely to have major effects on basic personality.
Cohen also points to the common observation that siblings, even same sex siblings, are radically different, both in personality and intelligence (sibling IQ differences average 12 points, which is 70% of the 18 point difference among children randomly selected from the same population). This is hard to explain by traditional social rearing theories since parents usually raise their children in the same way. Yet it is easily explained by a genetic theory (perhaps supplemented by there being a random component in development, such as Miller 1997 has proposed).
Many readers at this point may be puzzled by the idea that genetic effects can explain differences between siblings, since they think of genetic effects as making children just like parents. However, they are wrong as Cohen explains. When a new child is created, roughly half of the genes from each parent are passed on to the child in a random process. Each time a new sibling is created, this genetic lottery takes place. The result is that each child is genetically different. This is an important point that is not widely understood. Genetic theory predicts that siblings will be very different, while small sibling differences are predicted by the theory that parents shape their children.
The book (after an introduction) is divided into two parts. The first part focuses on how weak the case for parental effects on children's development is. The second presents the positive case for genetic and for pre-natal effects. If I had been writing the book I might have reversed the order, first documenting the cases where genetics and other non-parental causes were known to have an effect. After documenting how important genetic and pre-natal effects were, one would then have laid a solid basis for critiquing those who implicitly assumed these effects were non-existent, and then blamed the parents for their children's problems. This article will take the alternative approach.
Some of best-documented genetic effects are for intelligence. It is found that on IQ tests identical twins correlate about .85, fraternal twins about .60. Both sets of twins share the same home environments, but identical twins have all their genes in common while fraternal twins have only half of their genes in common. Thus, the extra. 25 correlation is the genetic effect of an extra .5 of genes in common. Thus, the total genetic effect should be twice .25, or .50. Thus, by the well established twin method, it is estimated that half of the variance in childhood intelligence is due to genetic effects (several .
Another method is to look at identical twins raised apart. These have the same genes but different environments. Five such studies (in three different countries) show an average correlation of .75. That identical twins raised apart come out so similar is powerful evidence for the power of genes. Supporting this interpretation is the fact that for studies of raised together twins when about 18 years old, the correlation for identicals is roughly the same as for younger identicals, but the fraternal correlation has dropped to about .4, again suggesting a higher heritability for adult intelligence.
How do you reconcile the higher heritabilities obtained from the identical twins raised apart methodology with the somewhat lower heritability from the standard twin method? Cohen (like others) reconciles these by noting that the studies of twins raised apart are of adults, while the studies of twins raised together are of children.
Their conclusion is that the heritability increases over time. To some this is surprising. One might expect the environmental effects to increase with time, causing them to be greatest in the adults. However, others would expect the influence of the common rearing by the parents to decrease over time. When children are young, the parents control the whole environment. As children age, they more and more pick their environments. Genes play a role in which environments they pick. All over the world teenagers are choosing friends, and even mates, that their parents disapprove of. A person who is smart and who enjoys intellectual activities will engage in intellectual activities, read books etc. By adulthood individuals are picking their own environments and the influence of the family is fairly minor.
Behavior genetics makes it possible to partition the environmental influence into that due to being reared in the same family (referred to as shared variance) and that due to other causes. One of the most striking findings is that the familial influences are so weak. Identical twins raised apart correlate about .75, while those reared together correlate .85. This suggests the effect of being reared together given identical genes is only .1, really quite small. As small as this is, studies of the IQ resemblance of unrelated children adopted into the same household show that they have essentially zero IQ resemblance. This suggests that the effect of being reared in the same family is essentially zero. Most of the environmental effects documented (and behavior genetic methods do prove that there are environmental effects) appear to be due to things that siblings within the same family do not share. (Rowe 1994 presents much evidence for this).
Supporting evidence for the role of genes is provided by many findings. Brain size correlates with IQ, tests done at 6 months (before most environmental influences have had a chance to exercise their effects) correlate with childhood IQ, adult IQ correlates with the biological relatives of adoptees, but not with that of the adoptive parents, and IQ correlates with reaction times. For more on these correlations see Miller's (1999) review and summary of Jensen's (1998) recent book, or the review and summary (Miller 1996) of Brand's book (Brand, 1996) with the same name. Interestingly, the publisher of Cohen's book is the same Wiley that withdrew Brand's book from the book stores, apparently because it discussed scientific facts about intelligence they did not want discussed.
The argument then goes on to discuss various aspects of personality, summarizing what is known about their biological causes. It is pointed out that where one identical twin commits suicide, in 15% of the cases the other twin also does so. While this may not sound like much, it implies that the rate is elevated by a factor of 800. For adaptively reared people who commit suicide, their biological, but not their adoptive relatives, have elevated suicide rates. Suicide and depressive disorders are found to run in families. In a study of Amish, 24 out of 26 suicides were associated with depressive disorders, although evidence is that only 20% of the depressed people try to kill themselves.
Crime is another thing parents often get blamed for. Yet consider the findings when adoptees whose biological mothers were felons were compared with adoptees whose mothers were non-criminal. The former were more likely to have been arrested (15% versus 2%), convicted (13% versus 1%) and incarcerated (10% versus 0%), as well as to have been diagnosed with antisocial personality. In Denmark, Medick's work showed that among adoptees, the number of convictions by the biological parents helped predict the probability that the adoptee would have a property crime conviction. The probability rose from 10% when the biological parent had no convictions to 20% when he had three or more convictions. No such relationship was found for the adoptees and their adoptive parents. Yet surely if criminality were related to something learned from the parents, a relationship would show up between the criminal behavior of adoptees and the adoptee's parents.
Very likely what is inherited is certain personality traits, including aggressivity. One study of adoptees that had an aggression disorder found that 30% of their biological parents had an antisocial personality disorder (but none of the adopted parents). Interestingly, a 30% rate is also found for non-adopted children and their parents.
The twin literature also shows a genetic effect, with an average of 10 studies showing the concordance rate for identicals to be 51%, and for fraternals a much lower 21%. Especially interesting is a study of eight identical twin -adoptees, each with a criminal record. Four, or fifty percent, also had a co-twin with a criminal record.
Interestingly, juvenile delinquency seems to show a lower heritability (but still a positive one). Pooled results for 7 twin studies showed a 84% concordance rate for identicals, versus a 69% rate for fraternals. Apparently, environmental factors such as peer pressure play a greater role in explaining the tendency of twins to be alike in delinquency. Evidence is presented that criminal tends and delinquents tend to be of low intelligence, so part of the heritability of criminality is a reflection of the heritability of intelligence.
Many parents are concerned with the combination of distractibility, impulsitivity, and hyperactivity known as attention deficit-hyperactivity disorder. This turns out to be related to having an antisocial personality and criminality. The biological fathers of hyperactive adoptee children were eight times as likely to have an antisocial personality. Twin studies show hyperactivity to have a high heritability, perhaps as high as 80%.
The natural question is where within the brain are the structures that affect these behaviors. A tumor affecting the part of the brain known as the hypothalmus is one source of violent anti-social behavior. Other neurological causes include an abnormality within the temporal lobes, or within the frontal lobes.
While twin and adoption studies are very powerful for demonstrating genetic effects, direct evidence of a particular gene's effects are also powerful evidence. Cohen reviews the evidence for many single gene effects, starting with Alzheimer's (admittedly a disease seldom blamed on poor parenting). While three single genes are known to cause early onset Alzheimer's, for the much more common late onset form, the research has focused on the gene that affects the E form of the apoliptoprotein. In one study, those with none of the E-4 form had a 20% Alzheimer's rate, 47% in those with one, and 91% for those with two.
Since one effect of Alzheimer's is to greatly lower intelligence, a widely disseminated gene that affects intelligence has now been identified. There is now evidence that compared with other Alzheimer's patients, those with a double dose of the E-4 allele have more of the amyloid plaques that are associated with the disease.
Cohen reports on some fascinating work on biological causes for behavior (i.e. causes apparently not due to differential parenting). Evidence is presented that low serotonin accompanies anti-social behavior in humans, including impulsive violence, impulsive fire setting, etc. One study followed aggressive children over a two-year period. Those with the lowest initial levels of serotonin were most likely to get into trouble. Survivors of an initial suicide attempt with low serotonin were more likely to make a follow-up attempt. Also, released criminals convicted of manslaughter or arson were followed up for three years. Those with the lowest serotonin activity were most likely to commit a violent crime.
Finally, there is evidence that for fraternal twins with an E4 allele, but without Alzheimer's symptoms, perform more poorly than their E4-free co-twins. As an aside this means that in this age group, a genetic polymorphism has been discovered that affect IQ. Although Cohen doesn't point it out, since the frequency of the E-4 Allele is known to differ between populations, the first gene producing racial differences in IQ has been identified.
Cohen mentions many fascinating findings. Wolfram's syndrome is a rare genetic condition associated with blindness, deafness and other conditions. About two thirds also have psychiatric conditons. It is a recessive condition requiring two copies of gene on chromosome 4 and affects about 1 in a 100,000 persons. However, single gene carriers have an increased risk (in one study 4-5 times greater-than-expected risk) of psychiatric behavior. Two studies have found such symptoms, and suicidal behavior, in biological relatives of Wolfram patients. Because one randomly picked from the population has a one in a hundred chance of carrying a Wolfram mutation, the total number of cases caused by this mutation is potentially huge.
Fragile X syndrome is a fascinating condition in which a large number of repeats on the X chromosome cause mental retardation. A recent discovery is that the number of repeats can increase over the generations, typically in the ovarian cells that mature into a woman's eggs (contrary to the general rule that genes are transmitted unaltered between the generations). Most individuals have a small number of repeats, (<50) that are normal and stable, but with 54-200 repeats there is a premutation which is likely to expand into a full mutation (with expansion from 90 repeats being virtually certain). One characteristic of this is that identical twins can differ in the number of repeats, making it possible for one genetic twin to exhibit genetic mental retardation, but not the other one. Cohen points out that this implies that not all of the differences found between identical twins are necessarily environmental in origin. He speculates that for schizophrenia and panic disorders, where the age of onset seems to drop over the generations, that the observed differences between identical twins could be caused by such a genetic mechanism.
Cohen claims that identical twins can even differ in sex. This happens when an XY embryo (which would imply a male) is created, but divides into two, creating identical twins. However, apparently in some cases in the next cell division, the Y chromosome has been known to get lost, creating individuals with an X but no Y, an abnormal arrangement called XO. This condition is known as Turner's syndrome. The children are females, look female and act very feminine, but because they lack ovaries, is infertile. In rare cases this affects only one of a pair of otherwise identical twins, producing identical twins of different sexes.
Another fascinating fact about Turner's syndrome females has recently been discovered. Turner syndrome females are often relatively impaired in social cognition, often being socially disruptive (notice that social skills are often assumed to be determined by rearing). Compared to Turner girls who received their single X chromosome from their father, those who received their single X chromosome from their mothers are more impaired academically (40% versus 16%) and socially (72% versus 29%). They are more likely to lack awareness of other's feelings, to be excessively demanding of other's time, to be difficult to reason with when upset, to be insensitive to other's body language, and to be unaware of what is socially appropriate.
This is believed to be an example of genomic imprinting in which the gene from one parent is apparently switched off. It is possible that genes for social cognition and self-control are actually switched off in the offspring when inherited from the mother. Cohen speculates that this might explain why males, who inherit only a single X chromosome from their mother (if they inherit a X chromosome from their father they become females) are socially less adroit than females. The females also inherit a X chromosome from their father, and this X chromosome presumably carries the relevant genes which are not switched off. This is just a theory, but a fascinating one.
Cohen provides a fascinating account of the biology of sex differences, presenting strong evidence that many of the sex differences are indeed biological. He discusses the fascinating Joan-John case in which one member of a set of identical twins had his penis accidentally damaged soon after birth. The doctors recommended that he be surgically altered to look female and be raised as a girl. The parents did this. For the first few years the sex reassignment appeared to have been a success. At this point, the "experiment" appeared to have demonstrated that social conditioning was everything and that two genetically different individuals, one raised as a male and one as a female, would adopt the behavior's of their respective sexes. This was very politically correct, and the initial report was widely written up in textbooks.
However, the full story was not as depicted in the textbooks. The child was very unhappy as a girl, became depressed, and even attempted suicide, always feeling she was not really a girl. She rejected dresses, dolls, and any attempt to get her to imitate her mother's make up rituals. She preferred to imitate her father shaving and to play with boys. She finally unilaterally adopted a masculine identity, deciding to live as a man, and quit trying to act as a girl. Finally her parents revealed her sex of birth. He (back in his male identity) then had a mastectomy to remove the hormonally created breasts, and an operation to recreate a penis. He eventually married a woman, and adopted her children. This sad story seems to show that certain parts of the brain are masculinized or feminized early, and not easily altered by environmental conditions. Six cases are known of XY males born without a penis, castrated at birth, and reared as girls. One researcher noted that all six were basically male in their minds and behavior.
While genes frequently explain our descendants' problems, sometimes the problems are due to prenatal environmental causes. Some twins are more like each other than others. An identical twin is created when the zygote (a fertilized egg) divides. If this occurs early, each twin develops its own placenta and may have different levels of nutrition. If it occurs late, the twins share a placenta and hence a blood supply. If the split occurs late the twins may be mirror image twins in which certain aspects are reversed, notably handiness. Reportedly, twins discordant for handiness are more likely to be concordant for schizophrenia (60% versus 32%). This is consistent with schizophrenia being caused by early viral infection, which would affect both twins that shared a placenta. Interestingly, identical twins with a significant difference in fingerprints (at the base of the index and third fingers) are more likely to be discordant for schizophrenia. This effect is believed to occur because the finger print difference indicates differences in nutrition or oxygen supply between the twins that could also affect brain development. Interestingly when one or both of the twins have unusually asymmetric fingerprints, there is an increased probability of severe schizophrenia. Interestingly, twin pairs with one member with such an asymmetry tend to be unusually different on certain traits including anxiety, depression, phobias, concerns about health and morals, and schizophrenia like peculiarities of thinking.
Cohen points to the experience of homosexuals as evidence of the power of biology, since most homosexuals adopt their sexual orientation and behavior in spite of strong social pressures not to be homosexual. This is one case where a naturally born child may feel like a stranger in the nest. He describes the evidence that there is a genetic influence on homosexuality, which comes from both twin studies and from research dealing with a specific location on the X chromosome. Twin studies suggest a heritability of between 25% and 75%. Specific markers on a region of the X chromosome known as X28 were shared in 33 out of 40 homosexual pairs of brothers, far above the expected 50% sharing if there was no genetic effect. Much of this evidence is also reviewed in a paper by the author of this review (Miller 1998), which also attempts to reconcile evidence for the heritability of homosexuality with the evolutionary argument that genes for homosexuality should be strongly selected against.
Especially interesting is Cohen's discussion of obsessive-compulsive disorder and Tourette's syndrome (a ticing disorder). Fifty percent of those afflicted with Tourette's and 20% of their relatives have obsessive-compulsive disorder. The 20% rate is found even for those Tourette's patients who themselves do not display obsessive-compulsive disorder.
Children often develop obsessive-compulsive disorder after recovering from a strep infection. The immune system of virtually all obsessive-compulsive disorder children, even those who have never had strep, are especially reactive to an antibody-a specific protein called D8/17. He suggests that obsessive-compulsive disorder may be an automimmune disease in which the region of the brain called the caudate is adversely affected. The same holds for Tourette's syndrome.
While this is basically a book for the layman (and especially for parents), the professional will learn things he had not been aware of. For instance, this writer was fascinated to learn that an adoption study had been conducted dealing with birth order. Various studies (Sulloway 1996) have shown that personality and adult behavior vary with birth order. For instance, the later born are more open to experience than the first-born. This has usually been explained by social theories in which children of various birth orders experience somewhat different environments. For instance, the later born usually have to deal with the presence of older siblings who are stronger than they are. A standard genetic theory cannot explain these birth order effects, since the probability of getting any particular gene is independent of birth order. However, there could be some biological effects that are neither social nor genetic in origin. Recently, University of Texas psychologists Jeremy Beer and Joe Horn used an adoption study to look for evidence of the social effects of rearing. They compared first-born adoptive children reared either as a social senior with a younger sibling, or as a social junior with an older sibling. None of the IQ and personality effects that had been reported in other studies showed up.
This points to a prenatal effect rather than a rearing effect. Blanchard and Klassen (1997) have proposed a biological theory that could explain the birth order effect on homosexuality. Miller (forthcomng) has extended this theory to explain other birth order effects.
One could go on and on with the evidence that biology affects many mental and psychological aspects of behavior, most of which have been attributed to various aspects of parenting. As noted, this reviewer would have presented this evidence earlier. Once one is aware of the positive evidence for genetic and prenatal effects, one becomes very aware of the faults in the many studies that the media and many textbooks present as showing the importance of environmental effects in general, and parenting in particular. The first major section of the book is devoted to critiquing the various studies that have claimed to demonstrate the importance of environmental effects, and the importance of parenting.
The most common example of overstating environmental effects is to simply ignore the possibility (or probability) that the observations are due to genetic effects. For instance a young child may readily exhibit anger, just as his parents do. Instead of saying that the child inherited a quick temper, it may be asserted that he never learned from his parents to control his temper.
Mistakes of this type are frequently made with regard to intelligence, and traits associated with intelligence such as academic and economic achievement. Intelligence is know to be strongly correlated with academic and economic achievement (see Herrnstein & Murray 1994 or Miller's 1995 review and summary of their book in this journal). As shown above there is substantial heritability for intelligence. Yet, when intelligent parents are observed to have intelligent children, the obvious explanation (to those familiar with the behavior genetic evidence) is rejected, namely that the children inherited the parents' intelligence. His problem solving skills are instead attributed to how his parents reared him. The same is often done for academic achievement, which in children is often attributed to the parents good rearing and the good environment provided, rather than to smart parents having smart children who do well in school.
Because intelligence (whether that of the parents or of the children) is often not measured in studies, while traits such as parent's education or occupation are, the child's good performance is often attributed to socioeconomic class rather than the indirect effects of inheritance of intelligence.
In considering the environment's influence, it should be recognized that people often select their own environments, and frequently their genetic traits affect the environment they select. For instance, smart, motivated students often select the harder courses in school, and go to Ivy League schools. It does not follow automatically that just because those who are educated at such places know more, that their knowing more is due to the superior education given there.
It should also be realized that people construct their own environments. A well-known measure of the quality of the home environment, the HOME inventory, proves to have substantial heritability. What is happening? Apparently many of the things measured are a response to the children's traits, which are themselves inherited. For instance, a home filled with books is more likely to be provided to a child who likes to read.
Another example is that the self reported numbers of spankings parents used on their 6-9 year olds correlated with the amount of lying, cheating and disruptive behavior their children displayed two year later in school. The newspaper suggested that parental use of corporal punishment actually made children more anti-social in later years. However, a very plausible alternative is that bad children get spanked, and children who are bad when young are bad later.
Cohen points to a study that showed that children with fewer behavioral problems tended to have more supportive parents, but the correlation was a modest .23. For adoptive parents the correlation was a much lower .07. Apparently the personality traits that make for supportive parents also make for well-behaved children, and the two are correlated primarily because parents and children have 50% of their genes in common.
At one point Cohen notes "When we are tempted to conclude, No wonder little Billy is so neurotic, just look at his parents, we should consider the reverse; No wonder his parents are so neurotic, just look at little Billy."
By such arguments Cohen builds up to his conclusions that while environmental influences and parenting are important, their impact is limited. Parents should not blame themselves when their offspring indeed turn out to be "strangers in the nest."
References
Blanchard, R., and Klassen, P. 1997 H-Y antigen and homosexuality in men. Journal of Theoretical Biology, 185: 373-378.
Brand, Christopher 1996 The G Factor: General intelligence and its implications. Wiley UK
Herrnstein, R. J. & Murray, C. 1994 The Bell Curve: Intelligence and Class Structure in American Life. New York: The Free Press.
Jensen, Arthur, 1998 The g Factor: The Science of Mental Ability. New York: Greenwood.
Miller, Edward M 1995 Race, Socioeconomic Variables, and Intelligence: A Review and Extension of The Bell Curve, Mankind Quarterly, Vol. XXXV, (Spring), No. 3, 267-291.
Miller, Edward M, 1996 The g Factor: The Book and the Controversy. Journal of Social, Political, and Economic Studies, Vol. 21, (Summer) No. 2, 221-232.
Miller, Edward M, 1997 Could Nonshared Environmental Variance Have Evolved to Assure Diversification Through Randomness? Evolution and Human Behavior, Vol. 18 No. 3, 195-221.
Miller, Edward M, 1999. Intelligence, Income, Inequality, Mankind Quarterly, Vol. 39, (Spring) No.3 (forthcoming).
Miller, Edward M, forthcoming. Homosexuality, Birth Order, and Evolution: Towards a Equilibrium Reproductive Economics of Homosexuality, Archives of Sexual Behavior
Rowe, D. C. 1994 The Limits of Family Influence: Genes, Experience, and Behavior. New York: Guilford.
Sulloway, Frank J. 1996 Born to Rebel: Birth Order, Family Dynamics, and Creative Lives. New York: Pantheon Books.
