Behavior Genes
591
twins where the parents were correct in
their assumption about zygosity. DZ twins
thought to be MZ would be treated more
similarly than correctly assigned DZ twins.
In fact, studies of traits such as personality,
intelligence quotient (IQ), and depression
symptoms in childhood have found that
parental misclassiFcation has negligible
effect on twin resemblance. ±inally, stud-
ies of MZ twins who have been reared
apart make it possible to eliminate the
effects of shared environment altogether,
and comparisons of the MZ twins reared
together with those reared apart for traits
such as personality reveal that correlations
are similar for both types. Thus, even if MZ
twins do share a greater environment than
DZ twin pairs, it does not signiFcantly
influence their similarity for behavioral
phenotypes. Therefore, the available evi-
dence suggests that the equal environment
assumption is a reasonable conjecture. The
potential effects of prenatal environment
also need to be considered in twin studies
as MZ twins often share a chorion, whereas
DZ twins never do. This may also influ-
ence MZ twin similarity, although there is
little evidence to support this.
Twin studies also assume random mat-
ing of the parents, in which proband
parents do not correlate more for the ob-
served phenotype than would be expected
by chance. However, assortative mating
where the parents correlate highly for the
trait has been detected for some pheno-
types, such as IQ and antisocial behavior.
This may increase the proportion of pre-
disposing genes shared by DZ twins above
the expected 50%; thus DZ twins would be
phenotypically more similar than if ran-
dom mating was occurring. There is no
effect on the genetic relatedness between
MZ twins as they already share 100% of
their genes. Consequently, the difference
between MZ and DZ twin correlations will
be reduced, resulting in underestimation
of the heritability.
Despite their limitations, twin studies re-
main the most used tool in disentangling
the genetic and environmental influences
on behavior. ±igure 4 summarizes twin
study results for a number of disorders
and traits. There are clear MZ–DZ dif-
ferences for some phenotypes such as
schizophrenia, manic depressive disorder,
unipolar depression, and cognitive abil-
i
t
ya
sm
e
a
su
r
edb
yIQt
e
s
t
s
,sugg
e
s
t
ing
genetic effects. In the case of bulimic be-
havior and childhood conduct disorder,
there is evidence of familiality (there are
positive correlations) but little suggestion
of genetic effects (modest differences in
the MZ and DZ correlations).
However, the classic twin study is not
the end of the story. There are a num-
ber of possible directions in which twin
studies can be elaborated upon. Extend-
ing twin studies to include other family
members is one useful method that has
already been applied and which has poten-
tial for future research. ±urthermore, the
offspring of MZ twin pairs are of particular
interest. The son of an MZ father shares
50% of his genes with both his father and
his uncle, thus comparison of the corre-
lation between uncles and nephews with
the father and son pairs may inform us
more about the genetic basis of the pheno-
type, whilst eliminating any confounding
effects of common environment or G
×
E
correlation. Similarly, we could compare
both sets of offspring from such a family
set-up as these are genetically similar to
half sibs. See ±ig. 5.
2.3
Adoption Studies
The effects of genes and environment on
development of behavioral traits can also
previous page 591 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 593 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off