tions does not rule out a substantial influence of the environment on intellec-
tual differences. After all, people learn child-rearing practices and other skills
relevant to the cognitive development of the child from their parents. For ex-
ample, children may acquire an interest in reading from their parents and pass
this interest on to their own children. Children who become interested in read-
ing are likely to read more, get better at reading, and so do well on IQ tests that
are typically saturated with test items that depend on language skills.
TheimportanceoftheenvironmentinaccountingforfamilialIQcorrelations
is suggested by the fact that children and their parents are likely to grow up in
similar cultural and economic circumstances. Even adopted children are likely
to be placed in homes similar in educational and economic background to the
homes of the biological parents. Furthermore, children adopted as infants may
be more likely to have suffered from prenatal problems, which may undermine
their intellectual development and reduce the correlation between their IQ
scores and the IQ scores of the adopted parents.
Some of the familial correlations demonstrate an important effect of environ-
ment on intellectual differences. The IQ correlation between unrelated children
living together is about .3, which is certainly much greater than zero. So there is
at least some tendency for people who have dissimilar genes but similar family
backgrounds to have similar IQ scores. Also, the correlation between IQ scores
for ordinary biologically related siblings is about .4, which is much lower than
the correlation for fraternal twins reared together (about .6), even though the
genes of fraternal twins are no more similar than the genes of ordinary siblings.
Presumably, though, the family environments of fraternal twins are more simi-
lar than the family environments of ordinary siblings, because twins share the
same period of family history.
Turning to the twins-reared-apart paradigm, it is worth noting that the envi-
ronments of twins reared apart are not necessarily all that different from those
of siblings reared together. As I mentioned before, adoption agencies usually
try to place adoptees in homes similar to the home of the biological parents.
Furthermore, when twins are raised separately, one twin is often reared by an-
other family member; twins are not usually separated until later childhood; and
the twins often remain in contact with one another. In other words, there is a
kind of environmental ‘‘contamination’’ that may make the environmental in-
fluences on the twins reared apart more similar than is commonly appreciated.
Finally, twins are more susceptible to prenatal trauma, which can result in
mental retardation, reflected in lowered IQ scores for both twins, even if reared
apart. This inflates the IQ correlation between twins (see Anastasi, 1988).
Anotherkindofproblemwiththetwins-reared-apartparadigmisthatitdoes
not identify which shared genes are the underlying cause of the similarity in IQ
scores. One possibility is that the genes that produce the high correlations in-
fluence biological functions that are directly involved in many cognitive pro-
cesses. But there are other possibilities.
Consider the following hypothetical scenario. Identical twins share facial and
bodily features, the characteristics of which are established primarily by genes.
How people are treated depends to some extent on their physical appearance.
Consequently, people’s social skills, confidence, and so on depend to some ex-
796 R. Kim Guenther