To What Degree Does Environment Affect I.Q.?

The influence of environment on the expression of genetic traits is especially hard to study when a number of different genes affect the trait. Nowhere has this difficulty led to more controversy than in studies of I.Q. scores. I.Q. is a controversial measure of general intelligence based on a written test that many feel to be biased towards white middle-class America. However well or poorly I.Q. scores measure intelligence, a person’s I.Q. score has been believed for some time to be determined largely by that person’s genes.

How did science come to that conclusion? Scientists measure the degree to which genes influence a multi-gene trait by using an off-putting statistical measure called the variance. Variance is defined as the square of the standard deviation (a measure of the degree-of-scatter of a group of numbers around their mean value), and has the very desirable property of being additive — that is, the total variance is equal to the sum of the variances of the factors influencing it.

The degree of gene influence, or heritability, is defined as the fraction of the total variance that is genetic:

H = Variance (genes)/Variance (total)

What factors contribute to the total Variance? There are three. The first factor is variation at the gene level, some gene combinations leading to higher I.Q. scores than others. This is the term “Variance (genes)” in the simple equation above. The second factor is variation at the environmental level, some environments leading to higher I.Q. scores than others. This term is expressed as “Variance (environment).” The third factor is what a statistician calls the covariance, the degree to which variation in the value of one factor influences variation in the value of others. In this instance, covariance expresses the degree to which environment affects the impact of genetic differences in I.Q. The covariance term is expressed “Variance (covar).”

Now the heritability of I.Q. can be expressed in terms that can be measured:

H = [Variance(genes)]/[Variance(genes)+Variance (environment)+Variance (covar)]

The environmental contributions to variance in I.Q. can be measured by comparing the I.Q. scores of identical twins reared together with those reared apart (any differences should reflect environmental influences). The genetic contributions can be measured by comparing identical twins reared together (which are 100% genetically identical) with fraternal twins reared together (which are 50% genetically identical). Any differences should reflect genes, as twins share identical prenatal conditions in the womb and are raised in virtually identical environmental circumstances, so when traits are more commonly shared between identical twins than fraternal twins, the difference is likely genetic.
When these sorts of “twin studies” have been done in the past, researchers have uniformly reported that I.Q. is highly heritable, with values of H typically reported as being around 0.7 (a very high value). While it didn’t seem significant at the time, almost all the twins available for study over the years have come from middle-class or wealthy families.

The study of I.Q. has proven controversial, because I.Q. scores are often different when social and racial groups are compared. What is one to make of the observation that I.Q. scores of poor children measure lower as a group than do scores of children of middle-class and wealthy families? This difference has led to the controversial suggestion by some that early childhood assistance programs such as Head Start are not worthy of public support, as they can offer only limited help when genes are the determining influence.

What should we make of such a harsh conclusion? To make a judgement, we need to focus for a moment on the fact that these measures of the heritability of I.Q. have all made a critical assumption, one to which population geneticists, who specialize in these sorts of things, object strongly. The assumption is that environment does not affect the impact of gene differences on I.Q. — that is, that the covariance term in the above equation is zero.

Recent studies have allowed a direct assessment of this assumption. Importantly, it proves to be flat wrong.

In November of 2003, researchers reported an analysis of twin data from a study carried out in the late 1960s. The National Collaborative Prenatal Project, funded by the National Institutes of Health, enrolled nearly 50,000 pregnant women, most of them black and quite poor, in several major U.S. cities. Researchers collected abundant data, and gave the children I.Q. tests seven years later. Although not designed to study twins, this study was so big that many twins were born, 623 births. Seven years later, 320 of these pairs were located and given I.Q. tests. This thus constitutes a huge “twin study,” the first ever conducted of I.Q. among the poor.

When the data were analyzed, the results were unlike any ever reported. The heritability of I.Q. was different in different environments! The influence of genes on I.Q. was far less in conditions of poverty, where environmental limitations seem to block the expression of genetic potential. Specifically, for families of high socioeconomic status, H = 0.72, much as reported in previous studies, but for families raised in poverty, H = 0.10. The lower a child’s socioeconomic status, the less impact genes had on I.Q.

These data say that the genetic contributions to I.Q. don’t mean much in an impoverished environment. Clearly, improvements in the growing and learning environments of poor children can be expected to have a major impact on their I.Q. scores. Additionally, these data argue that the controversial differences reported in mean I.Q scores between racial groups may well reflect no more than poverty, and are no more inevitable.

©2003 Txtwriter Inc.

Learn More Related Articles Homepage