Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Myths in Psychology and Education Regarding the Gene-Environment Debate

by Robert J. Sternberg & Elena L. Grigorenko - 1999

This article describes seven myths regarding the gene-environment debate that permeate the thinking of many educators and psychologists. These myths can lead to serious mis-interpretations of the role of genes in behavior and to false conclusions regarding the modifiability of behavior. We examine each myth, dispel it, and state what we believe to be correct conclusions from different kinds of behavior-genetic and related data.

This article describes seven myths regarding the gene-environment debate that permeate the thinking of many educators and psychologists. These myths can lead to serious misinterpretations of the role of genes in behavior and to false conclusions regarding the modifiability of behavior. We examine each myth, dispel it, and state what we believe to be correct conclusions from different kinds of behavior-genetic and related data.

It is difficult to pick up a newspaper or magazine without finding a report of some newly discovered gene that is alleged to control such-and-such a trait. Psychologists and educators rightfully wonder what the implications of such findings are for their work. Their question is whether they ought somehow to be changing what they are doing if, indeed, certain abilities or disabilities, personality traits, or motivational sets are inherited. The goal of this article is to respond to this question. Whether abilities or disabilities, personality attributes, or motivational inclinations are inherited fully or in part has virtually no bearing on what educators should do in their classrooms. In particular, what we know about heritability has no bearing on how modifiable attributes are (e.g., learning skills, thinking skills, motivation to learn, and so on). Evidence suggests that environment can make a powerful difference on academic skills and performance, whatever the heritabilities of various abilities may be. From the standpoint of education and psychology practitioners, the heredity versus environment debate is an unnecessary detour. The rest of this article will be devoted to elaborating upon this argument. Another inquiry into similar issues as they relate to psychology and social policy can be found in Block (1995).

The article is organized in terms of seven myths many psychologists and educators believe about heredity and environment. After readers understand these myths they will see why the heredity-environment debate has been, and continues to be, a red herring—a distraction from issues that truly are important in psychology and education.


Heritability (also called h2) is the ratio of genetic variation to total variation in an attribute within a population. Trait variation in a population is referred to as phenotypic variation, whereas genetic variation in a population is referred to as genotypic variation. Thus heritability is a ratio of genotypic variation to phenotypic variation. Heritability has a complementary concept, the one of environmentality. Environmentality is a ratio of environmental variation to phenotypical variation. Note that both heritability and environmentality apply to populations, not to individuals. There is no way of estimating heritability for an individual, nor is the concept meaningful for individuals.

Heritability is typically expressed on a 0 to 1 scale, with a value of 0 indicating no heritability whatsoever and a value of 1 indicating complete heritability. Heritability and environmentality add to unity (assuming that the error variance related to measurement of the trait is blended into the environmental component). Heritability tells us the proportion of individual-difference variation in an attribute that is inherited within a population. Thus if IQ has a heritability of .50 within a certain population, then 50% of the variation in scores on the attribute within that population is due (in theory) to genetic influences. This statement is completely different from the statement that 50% of the attribute is inherited. Even experts in a given field (e.g., Perkins, 1995), however, can make the mistake of believing that heritability refers to a proportion of a trait rather than to a proportion of individual-differences variation in a trait.

The distinction is important. Consider, for example, the attribute of IQ. IQ is expressed on an interval scale, meaning that arithmetical differences are meaningful: In theory, the difference between an IQ of 80 and 90 is the same as the difference between an IQ of 110 and 120. Multiplicative differences are not meaningful, however: An IQ of 120 does not indicate twice as much intelligence as an IQ of 60. Multiplicative differences cannot be meaningful, because there is no known meaningful 0 point for intelligence, the latent attribute alleged to underlie IQ. We don’t know what zero intelligence would be. An IQ of 0 does not mean “no intelligence.”

As an analogy, the Fahrenheit or Celsius temperature scales are interval scales, because 0 degrees F or 0 degrees C does not mean “no temperature.” In contrast, the Kelvin temperature scale is a ratio scale, where 0 degrees K truly means no temperature. IQ, of course, is like the Fahrenheit and Celsius scales, not like the Kelvin scale. The analogy breaks down, however, because whereas we know what 0 would mean in terms of temperature, we do not know what 0 would mean in terms of intelligence. An extraterrestrial alien, or an infant, for example, might score 0 on an adult test of intelligence, but such a score would not indicate a total lack of intelligence. Likewise, standardized and teacher-made tests are on interval, not ratio scales. A score of 0 does not mean that the individual has no ability or that the individual knows nothing whatsoever about the subject matter taught.

An important implication of these facts is that heritability is not tantamount to genetic influence. An attribute could be highly genetically influenced and have little or no heritability. The reason is that heritability depends on the existence of individual differences. If there are no individual differences, there is no heritability (because there is a 0 in the denominator of the ratio of genetic to total trait variation in a given population).

For example, being born with two eyes is 100% under genetic control (except in the exceedingly rare case of someone with a genetic coding for one or no eyes, which we will not deal with here). Regardless of the environment into which one is born, a human being will have two eyes. But it is not meaningful to speak of the heritability of having two eyes, because there are no individual differences. Heritability is not 1: It is meaningless (there is 0 in the denominator of the ratio) and cannot be sensibly calculated.

Consider a second complementary example: occupational status. It is heritable (Plomin, DeFries, & McClearn, 1990), but certainly it is not under direct genetic control. Clearly there is no gene or set of genes for occupational status. How could it be heritable, then? Heredity can affect certain factors that in turn lead people to occupations of higher or lower status. Thus if things like intelligence, personality, and interpersonal attractiveness are under some degree of genetic control, then they may lead in turn to differences in occupational status. The effects of genes are at best indirect (Block, 1995). Other attributes, such as divorce, may show heritability but, again, they are not under direct genetic control.


When discussing the unequivocal importance of heritability studies, scientists usually state that this type of research provides evidence of the presence or absence of genetic influences on the variability of a given trait. This statement is absolutely correct. The heritability coefficient is not the only one that could be used to obtain such evidence, however. One example of a different indicator is lr. Let us assume that an individual A has a distinct behavioral condition, for example, attention deficit. He has trouble paying attention in class, tends to have trouble staying in his seat, and seems to lack focus in his work. Even when he pays attention his attention span is very brief. Knowing the prevalence (i.e., the overall frequency) of the condition in the studied population, we can calculate the probability that a relative B, a sibling of an individual A, might have the same condition. Based on this calculation we can define the risk ratio for all siblings of all affected individuals of having attention deficit as compared with population prevalence (Risch, 1990a, 1990b). The higher the sibling risk ratio, the higher the evidence of familial aggregation of a given trait. Analyzing risk ratios for different types of relatives allows us to accumulate evidence for the presence or absence of genetic influences.

The concern could be raised that the concept of risk ratio is not really appropriate for studying the variability in normal traits, whereas the well-known h2 appears to be ideal for decomposing “normal” variance. Although this concern seems fair at first glance, two considerations must be addressed. First, the majority of non-h2 estimates of familial aggregation were developed in the fields of genetic epidemiology (the field concerned with the etiology of various diseases in populations) and genetics and have not been introduced into the field of psychology, where the dominance of h2 has been unchallenged. The current state of the art in genetic epidemiology and genetics is such that researchers are primarily concerned with discovering genetic mechanisms of diseases rather than with studying normal variability in a given trait in the population. This does not mean, however, that either Risch’s lr or a number of other measures of familial aggregation could not also be used to study normal variability. Second, in contrast to heritability-based studies, Risch’s lr and a number of other measures link genes and behaviors by calculating the degree to which specific typed genetic markers are shared by a given pair of relatives.

The ultimate goal of behavior-genetic research is to point to specific genes whose functions result in specific behavioral manifestations. Alteration of their functions through chemical means might ultimately result in life-saving or life-improving interventions, such as, if it were possible, intervening chemically when one or more genes associated with reading disability were conclusively identified. Thus by providing a direct link between genes and studied traits these concepts bridge the gap between the estimated statistics and the biological or psychological realities that are otherwise separated in h2. Ultimately, understanding gene functions may help in education if we can modify behavior through chemical interventions.


Although we may read about “the heritability of IQ” (e.g., Herrnstein & Murray, 1994), there really is no single fixed value that represents any true, constant value for the heritability of IQ or anything else, as Herrnstein and Murray (1994) and most others in the field (e.g., Bouchard, 1997) recognize. Heritability depends on many factors, but the most important one is the range of environments. Because heritability represents a proportion of variation its value will depend on the amount of variation. As Herrnstein (1973) pointed out, if there were no variation in environments, heritability would be perfect, because there would be no other source of variation. If there is wide variation in environments, however, heritability is likely to decrease.

For example, let us assume that the genetic variance of a hypothetical trait in a given population is 5 on some arbitrary scale. The total variance of the trait in this population is 10. Correspondingly, heritability of the trait is 50% (5/10). Now, let us assume that this population experienced rapid environmental change (as, for example, is the case in Eastern Europe). The trait variation has increased (suppose it becomes 15), the environmental variation has increased (suppose it is now 10), but genetic variation has remained the same (it is still 5). It can easily be calculated that heritability in the new environmental conditions for the same trait has changed—it is now 33% (5/15).

When one speaks of heritability one needs to remember that genes always operate within environmental contexts. All genetic effects occur within a reaction range so that, inevitably, environment will be able to have differential effects on the same genetic structure. The reaction range is the range of phenotypes (observable effects of genes) that a given genotype (latent structure of genes) for any particular attribute can produce, given the interaction of environment with that genotype. For example, genotype sets a reaction range for the possible heights a person can attain. There are no pure genetic effects on behavior, as would be shown dramatically if a child were raised in a small closet with no stimulation. Genes express themselves through covariation and interaction with the environment, as discussed further later.

The extent of the reaction range may differ for different attributes, a phenomenon referred to as canalization. Canalization indicates the extent to which an attribute develops without respect to the environment (Waddington, 1957). An attribute is highly canalized—or channeled—if it will develop almost independently of environmental circumstances. It is weakly canalized if its development depends highly on environmental factors. Like water at the bottom of a deep ditch, a strongly canalized trait lies deeper in its channel, where it is less accessible to the influences of the outside world. Basic memory abilities, as required in recalling a list of unrelated words, appear to be highly canalized (Perlmutter & Lange, 1978), whereas interpersonal skills, like the ability to persuade others or to negotiate effectively, are probably less so (see Gardner, 1983). In general, more highly complex attributes, such as intellectual skills, tend to be less highly canalized than are attributes relevant to simpler cognitive and motor abilities, such as sitting down or standing up.


Because the value of the heritability statistic is relevant only to existing circumstances, it does not and cannot address a trait’s modifiability. A trait could have zero, moderate, or even total heritability and, in any of these conditions, be not at all, partially, or fully modifiable. The heritability statistic deals with correlations, whereas modifiability deals with mean effects. Correlations, however, are independent of score levels. For example, adding a constant to a set of scores will not affect the correlation of that set with another set of scores.

The importance of the difference between heritability and modifiability is hard to overstate, especially because some investigators have suggested that the moderate heritabilities of intelligence imply that attempts to modify intelligence may be in vain (Herrnstein & Murray, 1994). The heritability of intelligence is generally estimated to be somewhere between .4 and .8 (Herrnstein & Murray, 1994; Sternberg & Grigorenko, 1997a). It increases with age, presumably because important differential effects of early environment become less important with age (Plomin, 1997). But what implications does such heritability, or any heritability, have for modifiability?

Consider height as an example of the limitation of the heritability statistic in addressing modifiability. Height is highly heritable with a heritability of over .90. Yet height also is highly modifiable, as shown by the fact that average heights have risen dramatically throughout the past several generations.

As an even more extreme example, consider phenylketonuria (PKU). PKU is a genetically determined, recessive condition that arises due to a mutation in a single gene on chromosome 12 (with a heritability of 1), and yet its effects are highly modifiable. Feeding an infant with PKU a diet free of phenylalanine prevents the mental retardation that otherwise would become manifest. Note also that a type of mental retardation that once incorrectly was thought to be purely genetic is not. Rather, the mental retardation associated with PKU is the result of the interaction with an environment in which the infant ingests phenylalanine; take away the phenylalanine and you take away the mental retardation.

Note that the genetic endowment does not change: The infant still has a mutant gene causing phenylketonuria. What changes is the manifestation of its associated symptoms in the environment. Similarly, with intelligence, or any other trait, we cannot change (at least based on our knowledge today) the genetic structure underlying manifestations of intelligence, but we can change those manifestations, or expressions of genes, in the environment.

Environment has a powerful effect on many attributes, including cognitive abilities. These effects may interact with genetic structures, but they nevertheless can result in massive modifications of demonstrated cognitive or other performances. Perhaps the simplest and most potent demonstration of this fact is called the “Flynn effect” (Flynn, 1987, 1994; Neisser, 1998). The basic phenomenon of the Flynn effect is an increase in IQ throughout successive generations around the world during the past 30 years.

The effect is powerful, showing an increase in IQ of about 18 or more points per generation for tests of fluid intelligence (Cattell, 1971; Horn, 1994; Horn & Cattell, 1966). Moreover, the effect has been shown for all 14 nations for which full data are available, and for other nations for which partial data are available. Fluid tests include ones such as the Raven Progressive Matrices, which measure a person’s ability to cope effectively with relatively novel stimuli and the relations between them. The mean effect has been inexplicably greater for tests of fluid abilities than for tests of crystallized, or knowledge-based abilities, which show an increase less than half as large. Perhaps cultural changes are affecting more fluid than crystallized abilities. But if linearly extrapolated, the difference would suggest that a person at the 90th percentile on the Raven test in 1892 would score at the 5th percentile in 1992 (see Neisser, 1998).

This effect is most likely environmental because a successive stream of genetic changes of such dramatic magnitude could not have occurred and exerted strong influence in such a short period of time. Psychometric tests of intelligence indicate that environment must be exerting a powerful effect on intelligence, perhaps in interaction with genes; intelligence can be, and is being, modified. Several explanations of the Flynn effect have been given (see Neisser, 1998, for a summary of various explanations of the effect). Among these explanations are increased schooling, greater educational attainment of parents, increased scaffolding by parents toward their infants, and technology. The important message, though, is that the environment can have and has had a massive effect on intellectual ability, regardless of what the heritability of intelligence may be. Unfortunately, we do not yet well understand these effects.

There are other puzzling effects as well. For example, we know that most of the environmental variation that matters to intelligence (and personality) is within rather than between families (Plomin, 1997). In other words, to the extent that environment matters, it seems to be largely because of differences in the way children are treated and react to their treatment within families rather than to differences across families. But why within-family effects should be so powerful is unknown. At the same time, we know that a single teacher can have very different effects on different students within the classroom because of the “fit” between student and teacher (see Sternberg, 1997; Sternberg & Grigorenko, 1995). So perhaps it should not be so surprising that a given family can affect different siblings differentially.

The Flynn effect is an intergenerational effect, and thus applies across individuals. But environmental effects within individuals also have been shown to be substantial. Evidence in support of environmental effects comes from two sources: natural experiments and intervention studies.

Adoption studies can provide natural experiments. Plomin and DeFries (1985) summarize the literature from the Colorado Adoption Project, which suggests that both heredity and environment have substantial effects on individual differences in intelligence. Two major adoption studies have been conducted in orphanages. One study was conducted by Dennis (1973) in Iran and the other study by Rutter (1996) in Romania. Dennis found that children placed in Iranian orphanages had low IQs. Probably because they were reared in institutions of different quality, girls had a mean IQ of about 50 whereas boys had a mean IQ of about 80.

Children adopted out of an Iranian orphanage by the age of 2 had IQs that averaged 100 during later childhood; they were able to overcome the effects of early deprivation. Children adopted after the age of 2 showed normal intellectual development from that point, but never overcame the effects of early deprivation; they remained mentally retarded. These results suggest that interventions to foster cognitive development need to start as early as possible.

Rutter’s Romanian project showed increases in mean IQ from 60 to 109 for orphans who came to the United Kingdom before 6 months of age. These children showed complete recovery from early mental retardation. Those who came to the United Kingdom after 6 months showed, on average, continuing deficits. This finding again argues for early interventions.

The second source of evidence in support of environmental effects is interventions (see review in Sternberg & Grigorenko, 1997b). Numerous studies of intervention programs show that such interventions can have at least some effect on cognitive abilities (e.g., Gutelius et al., 1972; Herrnstein et al., 1986; Johnson & Walker, 1991; McKey et al., 1985; Ramey, 1994). Thus these studies, too, show that environmental interventions can have an effect upon cognitive abilities, regardless of what the heritability of such abilities may be.


Those who seek to oppose heredity and environment and somehow find the exact percentages of variance attributable to each in various attributes are pursuing a fantasy. Genes and the environment typically work in tandem, not in opposition. When heritability coefficients are estimated they often include within them effects that are due both to covariance and to interaction of genes with the environment.

Consider first gene-environment covariance. Often the effects of genes and environments covary (or correlate)—in other words, they work together to yield results. There are three ways in which genetic and environmental forces can work in tandem (Plomin, 1994; Plomin, DeFries, & Loehlin, 1977).


In this case, children receive genotypes that are correlated with their family environment. For example, Nancy inherits musical abilities and then grows up in an environment where her musically oriented parents provide many musical opportunities.


In this case, people react to children on the basis of their genetic predispositions. For example, Kim-Li inherits musical abilities, and although her parents are not particularly musically oriented they notice that Kim-Li is. They decide, as a result of Kim-Li’s obvious talent, to offer her music lessons, although Kim-Li has not specifically requested such lessons.


In this case, children seek or create environments conducive to the development of their genetic predispositions. For example, Delphine inherits musical abilities and begs her parents for music lessons, which they then provide.

Notice how, in all three cases, it is quite difficult to separate the effects of genes from the effects of environment. Nancy, Kim-Li, and Delphine all have inherited musical talent. But in each case the environment is absolutely crucial to the development of this talent. What differs across the three kinds of covariation is the mechanism by which the genes and the environment relate to each other. Gene-environment covariance may be part of the reason children in the same family can have very different characteristics.

The second mechanism by which genes and environment work in tandem is through gene-environment interaction. In this case, environment affects the expression of genes, but affects this expression in different ways at different points along a continuum of genetic effect. For example, providing intensive musical lessons may help the child born with a high level of musical talent to become a world-class musician. The same musical intervention may have little or no effect on a child born with a very low level of musical talent.

Three main methods are used in attempts to separate effects of heredity and environment from each other. All three methods prove to be problematical in terms of their leading to clean separations of the two kinds of effects.

1. Family studies.

In family studies (also called “kinship studies”), the target of investigation is the degree to which different members of a family resemble each other as a function of their degree of relationship. These studies proceed by determining the proportion of genes different members of families should be expected to share, and then looking at their resemblance on a variety of attributes. For example, father and son would be expected to share half their genes, siblings would also share half their genes, but cousins would share fewer. If genes matter, then the greater the overlap in genes, the greater should be the resemblance in attributes.

There is an obvious problem with family studies (Plomin, 1994), namely, that they do not really enable researchers to make a clean separation between environmental and genetic effects. For example, siblings share roughly half their genes, but they also, to some extent, share environments. Cousins by blood share genes, whereas those by marriage do not. But it is difficult to say to what extent the two kinds of cousins share environments, because there is no widely accepted scale for measuring the precise similarity of the various environments. Thus, family studies need to be supplemented by other kinds of studies.

2. Adoption studies.

Adoption studies investigate the extent to which adopted children have attributes that are similar to those of both their adoptive and their biological family members. For example, in such a study an adopted child may grow up with an adoptive mother with whom the child shares no genes; but the child also has a biological mother with whom the child shares roughly half her genes but no environmental influences. Greater resemblance to the biological family suggests greater influence of heredity, whereas greater resemblance to the adoptive family suggests greater influence of the environment.

An example of this kind of study is the Texas Adoption Project, a longitudinal study following several hundred adopted children, along with the adoptive parents and biological parents (Loehlin, Horn, & Willerman, 1997). This project has shown, for example, that effects of genes upon measured intelligence generally increase with age, whereas effects of environment upon measured intelligence generally decrease with age. In other words, contrary to many people’s expectations, environmental effects become less rather than more important with age, at least in respect to intellectual abilities as measured by conventional tests.

Adoption studies are useful but never definitive. Although adopted children may not share the physical environment of the biological parents, neither are the adopted children randomly assigned to environments. As a result, there may end up being a covariation between the environments the children would have had with their biological parents and the environments that they have with their adopted parents; upper socioeconomic status (SES) children may be more likely to be placed in upper SES homes, for example, or children of a certain race or religion may be more likely to be placed in a family of the same race or religion.

3. Twin studies.

Probably the most powerful method for studying effects of heredity versus environment is through twin studies and, in particular, the study of identical twins raised apart. Occasionally, for one reason or another, identical twins are separated at or near birth and are then raised in different environments. They thus are placed in a situation where, in theory, they share no environment; but they nevertheless share all their genes. With this situation, it should be possible in theory to study cleanly the effects of genes and environment.

Although many studies of identical twins raised apart have been conducted, the most exhaustive is the Minnesota Study of Twins Reared Apart (Bouchard, 1997; Bouchard et al., 1990). This study, like the Texas study, found decreasing effects of environment and increasing effects of genes over time. For example, biologically unrelated children raised together showed a correlation in IQ of about .25 in childhood, but of –.01 in adulthood.

In practice, however, studies of identical twins raised apart are not quite as definitive as they might appear. For one thing, twins all have shared one common environment—the intrauterine environment—where influences, such as maternal alcohol consumption, may have occurred that affected the twins in common. For another thing, twins are typically not separated at birth, so that they spend at least some time growing up together. Finally, twins are usually not randomly placed into environments, but rather, are placed in environments that are somewhat similar, so that there is at least some correlation across environments.

All three areas of overlap in environment make it difficult to assign responsibility for any attribute entirely to genetics. Nevertheless, the studies of identical twins reared apart converge with other sources of evidence to suggest that genes have more of an effect on various kinds of development than we had thought, even as recently as five to ten years ago.


In our view, one of the worst offenses that has been committed by investigators of heredity and environment (or rather, most often, by interpreters of findings on heredity and environment) is that of generalizing the effects of within-population studies between populations. For example, some investigators have made attributions about effects of racial or ethnic group differences on the basis of behavior-genetic studies using the designs described above (Herrnstein & Murray, 1994), sometimes despite admitting that such conclusions are flawed. All of the behavior-genetic designs described above can ascertain effects of inheritance only within populations. They say nothing about sources of between-population differences.

An illustration of the impossibility of making between-population claims from within-population data has been given by Lewontin (1982). Here is one variant of it (see Herrnstein & Murray, 1994): Suppose one has a handful of corn seeds. Each seed is the repository of a large number of genetic attributes for the corn. If all the seeds are planted in the same environment, the heritability of attributes will reflect the restriction of range in environments. Given exactly the same environment, there is no differential effect of environment, and hence heritability is likely to be 1 or close to 1.

Now suppose, however, that half the seeds are placed in the fertile soil of an Iowa cornfield, and the other half are placed in the arid sand of the Mojave Desert. The difference in environments is enormous. The seeds planted in Iowa flourish; those planted in the Mojave Desert yield stunted and deformed plants. Note that even with identical genetic makeups and even if heritability of traits were 1, the results for the seeds could be quite different. One could not attribute to heritability the difference between the corn plants, which are from the same parent plant but placed in the two locations. Regardless of the heritability of the attributes within a given location, no conclusion could be drawn about inheritance from differential results across the two locations of planting.

Similarly, different populations—racial, ethnic, religious, or whatever—may encounter quite different environments, on average. Whatever the heritability of intelligence or other attributes within a given setting, no conclusions can be drawn about heritabililty as a source of differences across settings. The fact that IQs have increased so much over the years, as discussed above, suggests that environments differ widely over time. They likely differ substantially as well for members of different groups at a given time.

Nisbett (1995) has, in fact, reviewed published studies investigating sources of differences in cognitive abilities between White and Black individuals. These studies, using designs unlike the behavior-genetic studies described above, have directly sought to investigate genetic and environmental effects on intelligence. For example, one design has been to look at Black children adopted by White parents. Of 7 studies that have been published, 6 have supported primarily environmental interpretations of group differences, and only 1 study has not; the results of this one study (Scarr & Weinberg, 1976, 1983) are equivocal.

Moreover, there is much published evidence indicating that heritability estimates vary across populations. For example, estimates of the heritability of IQ in Russian twin studies conducted in the Soviet era tended to be higher than comparable estimates in the U.S. (Egorova, 1988; Grigorenko, 1990; Iskol’sky, 1988). This observation made sense—the environmental variation in Russia under the Soviet regime was constrained; consequently, the heritability estimates were higher. Most of the IQ heritability studies up to today have been carried out in various countries of the developed world. Relatively little information exists regarding the heritability of IQ in the developing world, although some studies suggest that heritability may be substantial, at least outside the Western countries that most often have been studied (Bratko, 1996; Lynn & Hattori, 1990; Nathwar & Puri, 1995; Pal, Shyam, & Singh, 1997). Obviously, without even knowing much about estimates of the heritability of IQ in different populations, we cannot speculate at this point about differences across these populations.


One of the most exciting realizations of the last decade is the enormous plasticity of human ontogenic development. We have learned that a given genetic makeup provides an individual with a certain likelihood of manifesting a behavior or a trait, but this likelihood varies as a complex function of a whole range of factors. Among these factors are whether the trait of interest is influenced by one gene (as in the case of phenylketonuria) or many genes (as appears to be the case for most psychological traits), whether the expression of this gene (or genes) is alterable by any environmental factors, and whether we understand the biological mechanism of the trait of interest to a degree that enables us to alter it.

First, it is the rule rather than the exception that human traits are extremely heterogeneous in their etiology and that nature has developed a number of mechanisms that can either lead to or prevent a manifestation of a given trait. In other words, different genetic mechanisms can result in the manifestation of the same phenotype (e.g., genetic forms of insulin-dependent diabetes mellitus) and the same genetic mechanism can lead to different phenotypic manifestations (e.g., a so-called statistic of penetrance shows, in a population, how many people who have a particular gene developed a corresponding phenotype). When the system is not completely deterministic, there is always room for intervention. The only issue is whether we know how to intervene effectively.

Second, there are no direct connections between genes and behavior. The more we learn about links between genes and behavior, the more we appreciate the role of the brain as the moderator of these links. The development of the brain and its organization are outcomes of continuous interactions between genetically coded programs for the formation and connectivity of brain structure and environmental modifiers of these codes (Fox, Calkins, & Bell, 1994). Understanding the underlying biological mechanism of behavior (or at least some components of it) means understanding complex multilevel biochemical pathways that lead to a manifestation of the trait.

Consider the following example: Alzheimer’s dementia, at least to some degree, is considered to be a familial condition—a genetically transmitted disorder. Recently it has been discovered that as much as a quarter to a third of brain tissue affected by the disease manifests plaque-like structures (lesions) containing a specific protein, AMY117 (Schmidt, Lee, Forman, Chiu, & Trojanowski, 1997). The production of this protein is controlled by an unknown gene (or set of genes). The hypothesis here is that it is not the gene itself that leads to the development of Alzheimer’s—rather, the protein that is produced by the gene somehow interacts with the brain at the locations of the lesions. And it is most likely the presence of the lesions that determines the cognitive profile of Alzheimer’s patients. The next step in understanding the biological mechanism of the formation of these lesions and the biological significance of having them in the brain is to find in the genome, and then clone, the gene(s) responsible for the production of this protein. The ultimate goal here is to understand the mechanism of AMY117 production and its functioning in the brain so that a cure can be developed. And this cure might turn out to be either pharmacological or dietary, as in the case of phenylketonuria.

Third, even if the genes are cloned and the mechanisms of their protein production are determined, there is still much variability in the system, given the way genes express themselves in a living organism. We are at the very beginning of understanding how expression of genes is regulated. For example, evidence suggests that certain variations in infants’ early social environments influence the shaping of the biological system involving the frontal lobe—the system that is associated with the expression and regulation of emotion (Dawson, Hessl, & Frey, 1994). At the same time, it has been shown that electrophysiological kindling (i.e., repeated electrostimulation of the animal brain that, when repeated at a given frequency, duration, and intensity, leads to the appearance of full-blown seizure in response to previously ineffective stimulus) and behavioral sensitization to psychomotor stimulants and stress provide some first clues into how repeated acute events can lead to extreme behavioral reactions.

Researchers argue that such repeated acute events can leave neurobiological residues in cells that, in turn, can alter gene expression and account for the observed long-lasting alterations in behavioral responsivity (Post, Weiss, & Leverich, 1994). In other words, environmental forces may change gene expression and, moreover, these forces might not necessarily be of a pharmacological nature. Human emotions, for example, can create specific biochemical environments in cells that might then change patterns of gene expression. In turn, change in the expression of genes might lead to a cascade of events, of both global and local impacts, among which might be neuronal regulation. For example, the research of Goodman and colleagues (Davis, Schuster, & Goodman, 1996; Schuster, Davis, Fetter, & Goodman, 1996a, 1996b) is concerned with the question of how changes in gene expression in a neuron with many axons and many synaptic connections can alter the strength of only some of the neuron’s synapses. Goodman et al.’s studies suggest that the nucleus of the neuron manages the assembly of the synaptic structure, while local biochemical factors in individual axons determine where in the nerve cell the synaptic structure gets placed. In other words, we are far from understanding the biological mechanisms of human development, but disjointed pieces of evidence suggest linkages between human environments, modified expressions of genes, and modified human behaviors.

In sum, it is an accepted belief among scientists today that genes determine the likelihood of behaviors, not the behaviors themselves. Heritability estimates do not explain the genetic regulation of behavior. Heritabilities are like snapshots of a dancer, which tell us neither what the dance is about nor what is coming next in the performance. The true genetic nature of humans is far from being defined, but what is absolutely clear is that genes do not act in a vacuum; they act in the environment and, because of that, when understood, their actions can be altered environmentally.


Our goal in this article has been to explore myths that have evolved regarding issues surrounding debates about the effects of genes and environment on various kinds of individual differences. We have not sought to take any particular position regarding the heritability of all the attributes that might be studied. Rather, we have tried to encourage readers to withhold judgment in the face of reports of studies that appear, on the surface, to assign definitive proportions of variation in attributes to heredity or environment. In general, there are so many factors that can affect these claims that the claims must be taken with a measured dose of skepticism.

We are not opposed to studies that investigate the behavior genetics of certain traits. Rather, our goal is to ensure that false claims made on the basis of such studies are identified by readers for what they are. Without an understanding of what the results of such studies truly mean, false conclusions about social and educational policies can be and have been drawn. At the same time, children’s futures can be compromised by our ignorance of what heritability statistics truly mean.

The principal implication of our review for the classroom is that the heritability of an attribute has no implication for the classroom. Heritability has nothing to do with modifiability. Teachers should be as diligent as possible in improving the intellectual and other academic skills of all their students, regardless of the literature on heritability. Moreover, group differences in levels of scores on tests should be understood in terms of the substantial power of the environment to make a difference. If the Flynn effect can result in massive gains in IQ over generations, then there are environmental forces to be mustered that can make a difference, even if at this time we only can speculate on what they are. One of the worst mistakes we can make in the classroom is to believe that we cannot make a difference, because once we believe it, it is likely to become a self-fulfilling prophecy. Wherever the answers may be to understanding the power of educational interventions, they most assuredly are not in studies of heritability.

The work reported herein was supported under the Javits Act program (Grant #R206 R50001) as administered by the Office of Educational Research and Improvement, U.S. Department of Education. The findings and opinions expressed in this report do not reflect the positions or policies of the Office of Educational Research and Improvement or the U.S. Department of Education. Requests for reprints should be sent to Robert J. Sternberg, Department of Psychology, Yale University, P.O. Box 208205, New Haven, CT 06520-8205.


Block, N. (1995). How heritability misleads about race. Cognition, 56, 99–128.

Bouchard, T. J., Jr. (1997). IQ similarity in twins reared apart: Findings and responses to critics. In R. J. Sternberg & E. L. Grigorenko (Eds.), Intelligence, heredity, and environment (pp. 126–160). New York: Cambridge University Press.

Bouchard, T. J., Jr., Lykken, D. T., McGue, M., Segal, N. L., & Tellegen, A. (1990). Sources of human psychological differences: The Minnesota study of twins reared apart. Science, 250, 223–228.

Bratko, D. (1996). Twin study of verbal and spatial abilities. Personality and Individual Differences, 21, 621–624.

Cattell, R. B. (1971). Abilities: Their structure, growth and action. Boston: Houghton Mifflin.

Davis, G. W., Schuster, C. M., & Goodman, C. S. (1996). Genetic dissection of structural and functional components of synaptic plasticity. Vol. 3, CREB is necessary for presynaptic functional plasticity. Neuron, 17, 669–679.

Dawson, G., Hessl, D., & Frey, K. (1994). Social influences on early developing biological and behavioral systems related to risk for affective disorders. Development and Psychopathology, 4, 759–780.

Dennis, W. (1973). Children of the creche. New York: Appleton-Century-Crofts.

Egorova, M. S. (1988). Genotip i sreda v variativnosti kognitivnykh phunktsii [Genotype and environment in the variation of cognitive functions]. In I. V. Ravich-Shcherbo (Ed.), Rol’ sredy I nasledstvennosti v formirovanii individual’nosti cheloveka (pp. 181–235). Moscow: Pedagogika.

Flynn, J. R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin, 101, 171–191.

Flynn, J. R. (1994). IQ gains over time. In R. J. Sternberg (Ed.), Encyclopedia of human intelligence (Vol. 1, pp. 617–623). New York: Macmillan.

Fox, N. A., Calkins, S. D., & Bell, M. A. (1994). Neural plasticity and development in the first two years of life: Evidence from cognitive and socioemotional domains of research. Development and Psychopathology, 4, 677–696.

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic.

Grigorenko, E. L. (1990). Esperimental’nor issledovanie protsessa vydvizheniia I proverki gipotez [Experimental study of hypothesis-making in the structure of cognitive activity]. Unpublished doctoral dissertation, NIOPP APN SSSR.

Gutelius, M. R., Kirsch, A. D., MacDonald, S., Brooks, M. E., McErlean, T., & Newcombe, C. (1972). Promising results from a cognitive stimulation program in infants: A preliminary report. Clinical Pediatrics, 11, 585–593.

Herrnstein, R. J. (1973). IQ in the meritocracy. Boston: Atlantic Monthly Press.

Herrnstein, R. J., & Murray, C. (1994). The bell curve. New York: Free Press.

Herrnstein, R. J., Nickerson, R. S., de Sanchez, M., & Swets, J. A. (1986). Teaching thinking skills. American Psychologist, 41, 1279–1289.

Horn, J. L. (1994). Theory of fluid and crystallized intelligence. In R. J. Sternberg (Ed.), The encyclopedia of human intelligence (Vol. 1, pp. 443–451). New York: Macmillan.

Horn, J. L., & Cattell, R. B. (1966). Refinement and test of the theory of fluid and crystallized intelligence. Journal of Educational Psychology, 57, 253–270.

Iskol’sky, N. V. (1988). Vliianie sotsial’no-psikhologicheskikh factorov na individual’nye osobennosti bliznetsov i ikh vnutriparnoe skhodstvo [The influence of social-psychological factors influencing twins’ individual characteristics and their similarity on psychological traits]. Unpublished doctoral dissertation, NIOPP APN SSSR.

Johnson, D. L., & Walker, T. (1991). A follow-up evaluation of the Houston Parent-Child Developmental Center: School Performance. Journal of Early Intervention, 15, 226–236.

Lewontin, R. (1982). Human diversity. New York: Freeman.

Loehlin, J. C., Horn, J. M., & Willerman L. (1997). Heredity, environment, and IQ in the Texas Adoption Project. In R. J. Sternberg & E. L. Grigorenko (Eds.), Intelligence, heredity and environment (pp. 105–125). New York: Cambridge University Press.

Lynn, R., & Hattori, K. (1990). The heritability of intelligence in Japan. Behavior Genetics, 20, 545–546.

McKey, R. H., Condelli, K., Granson, H., Barrett, B., McConkey, C., & Plants, M. (1985). The impact of Head Start on children, families and communities. Washington, DC: CSR.

Nathwar, S., & Puri, P. (1995). A comparative study of MZ and DZ twins on Level I and Level II mental abilities and personality. Journal of the Indian Academy of Applied Psychology, 21, 87–92.

Neisser, U. (Ed.). (1998). The rising curve. Washington, DC: American Psychological Association.

Nisbett, R. (1995). Race, IQ, and scientism. In S. Fraser (Ed.), The bell curve wars: Race, intelligence and the future of America (pp. 36–57). New York: Basic Books.

Pal, S., Shyam, R., & Singh, R. (1997). Genetic analysis of general intelligence “g”: A twin study. Personality and Individual Differences, 22, 779–780.

Perkins, D. N. (1995). Insight in minds and genes. In R. J. Sternberg & J. E. Davidson (Eds.), The nature of insight (pp. 495–534). Cambridge, MA: MIT Press.

Perlmutter, M., & Lange, G. (1978). A developmental analysis of recall-recognition distinctions. In P. A. Ornstein (Ed.), Memory development in children. Hillsdale, NJ: Erlbaum.

Plomin, R. (1994). The nature of nurture: Family environment. In R. Plomin (Ed.), Genetics, and experience: The interplay between nature and nurture (pp. 104–148). Beverly Hills, CA: Sage.

Plomin, R. (1997). Identifying genes for cognitive abilities and disabilities. In R. J. Sternberg & E. L. Grigorenko (Eds.), Intelligence, heredity, and environment (pp. 89–104). New York: Cambridge University Press.

Plomin, R., & DeFries, J. C. (1985). Origins of individual differences in infancy: The Colorado Adoption Project. New York: Academic Press.

Plomin, R., DeFries, J. C., & Loehlin, J. C. (1977). Genotype-environment interaction and correlation in the analysis of human behavior. Psychological Bulletin, 84, 309–322.

Plomin, R., DeFries, J. C., & McClearn, G. (1990). Behavioral genetics: A primer. New York: Freeman.

Post, R. M., Weiss, S. R. B., & Leverich, G. S. (1994). Recurrent affective disorder: Roots in developmental neurobiology and illness progression based on changes in gene expression. Development and Psychopathology, 4, 781–814.

Ramey, C. (1994). Abecedarian project. In R. J. Sternberg (Ed.), Encyclopedia of human intelligence (Vol. 1, pp. 1–3). New York: Macmillan.

Risch, N. (1990a). Linkage strategies for genetically complex traits. I: Multilocus models. American Journal of Human Genetics, 46, 222–228.

Risch, N. (1990b). Linkage strategies for genetically complex traits. II: The power of affected relative pairs. American Journal of Human Genetics, 46, 229–241.

Rutter, M. (1996, April). Profound early deprivation and later social relationship in early adoptees from Romanian orphanages followed at age 4. Paper presented at the 10th Biennial International Conference on Infant Studies, Providence, RI.

Scarr, S., & Weinberg, R. A. (1976). IQ test performance of black children adopted by white families. American Psychologist, 31, 726–739.

Scarr, S., & Weinberg, R. A. (1983). The Minnesota adoption studies: Genetic differences and malleability. Child Development, 54, 260–267.

Schmidt, M. L., Lee, V. M., Forman, M., Chiu, T. S., & Trojanowski, J. Q. (1997). Monoclonal antibodies to a 100-kd protein reveal abundant A beta-negative plaques throughout gray matter of Alzheimer’s disease brains. American Journal of Pathology, 151, 69–80.

Schuster, C. M., Davis, G. W., Fetter, R. D., & Goodman, C. S. (1996a). Genetic dissection of structural and functional components of synaptic plasticity. I: Fasciclin II controls synaptic stabilization and growth. Neuron, 17, 641–654.

Schuster, C. M., Davis, G. W., Fetter, R. D., & Goodman, C. S. (1996b). Genetic dissection of structural and functional components of synaptic plasticity. II: Fasciclin II controls presynaptic structural plasticity. Neuron, 17, 655–667.

Sternberg, R. J. (1997). Thinking styles. New York: Cambridge University Press.

Sternberg, R. J., & Grigorenko, E. L. (1995). Styles of thinking in school. European Journal of High Ability, 6(2), 1–18.

Sternberg, R. J., & Grigorenko, E. L. (Eds.). (1997a). Intelligence, heredity and environment. New York: Cambridge University Press.

Sternberg, R. J., & Grigorenko, E. L. (1997b). Interventions for cognitive development in children 0–3 years old. In M. E. Young (Ed.), Early child development programs: Investing in our children’s future (pp. 127–156). Amsterdam: Elsevier.

Waddington, C. H. (1957). The strategy of the genes. London: Allen & Unwin.

Cite This Article as: Teachers College Record Volume 100 Number 3, 1999, p. 536-553
https://www.tcrecord.org ID Number: 10322, Date Accessed: 10/24/2021 12:56:45 PM

Purchase Reprint Rights for this article or review
Article Tools
Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • Robert Sternberg
    Yale University
    E-mail Author
    Robert J. Sternberg is IBM Professor of Psychology and Education in the Department of Psychology, Yale University. Her is author of Successful Intelligence and of Beyond IQ. He is editor of Contemporary Psychology.
  • Elena Grigorenko
    Yale University
    Elena L. Grigorenko is Research Scientist in the Department of Psychology and Child Study Center at Yale University and Associate Professor of Psychology at Moscow State University. She is co-editor of Intelligence, Heredity, and Environment.
Member Center
In Print
This Month's Issue