The identification of factors that contribute to human intelligence has been a topic of study for over 100 years. Scientists, statisticians, educators, philosophers, and politicians have all weighed in on the matter. The U.S. eugenics movement of the early 1900s sought to prove that intelligence was almost entirely inherited. Today, it is well accepted that human intelligence is determined by a combination of genes and environment. Just what proportion of our intelligence is inherited remains a matter of scholarly debate, however.
Different studies have measured the heritability of IQ to be anywhere from 40% to 80%. If IQ were only 40% heritable, then only 40% of the variation in human IQ levels would be due to genetic factors. This would mean that environmental factors, such as schooling and diet, heavily influence human intelligence levels. On the other hand, studies that find the heritability of IQ to be closer to 80% suggest that environmental factors have little to do with the broad range of IQs observed in human populations. Such discrepancies as to the role of environment and genetics have led to the argument that ethnic differences in intelligence are due to genetic differences, thus fueling racism debates. These discrepancies have also led to suggestions that early intervention and schooling are a waste of resources. This long-lasting debate has motivated repeated scholarly attempts to ultimately define the heritability of IQ.
What Is Heritability and How Is It Estimated?
Figure 1: IQ is a continuous trait.
The IQ scoring system makes 100 the average value with different individuals' scores following a Gaussian distribution.
Copyright 2006 Alessio Damato.
Intelligence is a human trait that does not fall into distinct classes. This type of trait is known as a continuous trait, having a broad phenotype that is influenced by both genetic and environmental factors (Figure 1). Quantitative geneticists attempt to determine to what extent genetic and environmental factors influence the variations seen in continuous traits, like intelligence, in order to estimate the heritability of these traits. Heritability does not tell us how much of an individual's IQ level is innate and how much is learned. Rather, heritability is a statistical measurement that applies to an entire population. Specifically, it is a measure of the proportion of phenotypic variation observed in the population that can be attributed to genetic factors. Theoretically, heritability can have a value between 0 (indicating that none of the variation is due to genetic factors) and 1 (indicating that all of the variation in the trait is genetic).
To get some idea of how much of a trait is influenced by genes, you need to be able to compare how much the trait differs between people with similar genes. One way scientists are able to do this in humans is to use monogenic twins. Identical twins are useful for this purpose because they make it possible for scientists to compare genetically identical individuals for specific phenotypes, thus enabling them to quantify the heritability of certain traits. There is, however, the caveat that environment plays a large role in generating variation, even in genetically identical organisms.
How Do Researchers Use Twins to Estimate Heritability?
Monozygotic (MZ) twins, commonly known as identical twins, arise from the splitting of a single early embryo; therefore, these twins are genetically identical. On the other hand, dizygotic (DZ) twins, also called fraternal twins, arise from the simultaneous fertilization of two separate eggs. DZ twins are thus no more related genetically than full siblings, but they do share an identical prenatal environment. By measuring how well a set of twins correlates for a specific phenotypic measurement (such as IQ), researchers start to get an idea of how much of the trait is influenced by genes. If genetics alone accounted for all the variation in IQ scores, the MZ twin correlation would theoretically be 1.0. Similarly, siblings and DZ twins, who share half of their alleles, would have a correlation of 0.5 if IQ variation was due to genes alone.
Table 1: Correlations of IQ Among Differently Related Individuals-Observed and Expected Correlations if IQ Variation Were Due to Genes Alone
|
Relationship |
Reared |
Observed Correlation* |
Correlation if IQ Were Completely Heritable |
|
MZ twins |
Together |
0.85 |
1.0 |
|
MZ twins |
Apart |
0.74 |
1.0 |
|
DZ twins |
Together |
0.59 |
0.5 |
|
Siblings |
Together |
0.46 |
0.5 |
|
Siblings |
Apart |
0.24 |
0.5 |
|
Parent-child |
Together |
0.41 |
0.5 |
|
Parent-child |
Apart |
0.24 |
0.5 |
|
Adoptive parent-child |
Together |
0.2 |
0.0 |
*Data from Devlin et al., 1997, based on original tabulations by Bouchard & McGue, 1981.
However, as shown in Table 1, the observed correlation between MZ twins is not 1.0, so there must be environmental factors that influence IQ variation. The fact that MZ twins who were raised apart have a lower correlation than those raised together further demonstrates the influence of the environment on IQ variation.
Researchers typically attribute the phenotypic variance in a trait to a combination of three components. First, additive genetic factors (A) are the cumulative contributions of all the different alleles of the genes that influence the trait. Second, common environmental factors (C) are those environmental factors that are shared equally by both twins. Finally, unique environmental factors (E) are those environmental factors experienced by only one twin. With the assumption that MZ twins share 100% of their genetic sequence and 100% of their common environment (although the latter assumption is problematic), the MZ twin correlation (rMZ) is therefore equal to A + C. Similarly, under the assumption that DZ twins share half their genetic sequence and 100% of their common environment, then the DZ twin correlation (rDZ) is equal to ½A + C. These statements can be summarized as follows:
rMZ = A + C
rDZ = ½A + C
From these two equations, the values of A, C, and E can be derived:
A = 2(rMZ–rDZ)
C = rMZ–A
E = 1–rMZ.
To reiterate, heritability is the proportion of phenotypic variation due to genetic factors; in other words, it is equivalent to A. Thus, the heritability of a trait is calculated as twice the difference between the MZ and DZ twin correlations. Furthermore, the proportion of variance due to common environmental factors is equal to the MZ twin correlation minus heritability, and the proportion of variance due to unique environmental factors is the same as the amount by which the MZ twin correlation deviates from 1.
Different Interpretations from Different Twin Studies
After the genetic correlations are tabulated, researchers try to determine what types of environmental factors have the most influence on variation in the trait. Researchers consider common environmental effects experienced by each twin, as well as unique environmental effects. However, depending on the statistical modeling performed by the researchers, these different environmental effects can appear to have a greater or lesser influence on the overall phenotype. For example, studies reporting similar IQ heritabilities can differ markedly in their estimations of the influence of the prenatal environment, a factor that was largely ignored until a 1997 study by Devlin and colleagues.
In their study, Devlin et al. used data from 212 IQ correlations, then applied alternative models for environmental influence. They found that the data fit better when more consideration was given to maternal effects and shared fetal environment. These researchers argued that shared maternal environment may contribute more to correlations in IQ between twins than previously thought, and they therefore speculated that interventions aimed at improving the prenatal environment might lead to an overall increase in the population's IQ.
Heritability of Other Complex Human Traits
Quantitative geneticists are at the root of the "nature versus nurture" debate with their work to determine the genetic and environmental influences on many complex human traits. Researchers in this field have claimed a genetic basis for a wide range of disorders, including schizophrenia, depression, obesity, and cardiovascular disease, as well as other human traits such as personality attributes and aggressive or addictive behaviors. Evidence of a genetic basis for a disorder or behavior can motivate molecular biologists to track down susceptibility genes, while evidence for environmental causes of disease or negative behaviors can help foster environments that will promote better health.
With the myriad variables influencing complex human traits, each study in quantitative genetics apportions the effects of genes and environment slightly differently. Thus, scientists may never agree on the best way to stave off depression or increase a child's IQ because these issues are so complex.
References and Recommended Reading
Devlin, B., et al.
The heritability of IQ. Nature 388, 468–471 (1997) doi:10.1038/388468a0 (link to article)
Bouchard, T. J., Jr., & McGue, M. Familial studies of intelligence: A review. Science 212, 1055–1059 (1981)
