Since theorists have argued that temperament has biological roots, many studies have focused on genetic and neurological correlates of different behavioral styles. Most (if not all) temperament dimensions appear to be moderately heritable, with shyness showing the highest heritability. That is, the more closely people are genetically related, the more alike they are in temperament. Much of the evidence supporting this conclusion comes from twin studies, which compare the behavioral similarity of monozygotic (identical) twin pairs to dizygotic (fraternal) twin pairs. Monozygotic twins inherit identical genotypes because they develop from the same fertilized egg. In contrast, dizygotic twins inherit, on average, 50 percent of their segregating genes. If genetic differences across people are associated with temperament differences across people, then identical twin similarity should be twice as high as fraternal twin similarity. Plomin and others found that this is true for temperament characteristics such as emotionality, activity level, sociability, and shyness.
Interestingly, some studies show that identical twins are more than twice as similar as both fraternal twins and other pairs of relatives (e.g., parents and their children and biologically related siblings). How could this be? David Lykken and his colleagues hypothesized that temperament (and personality) differences are associated with genetic effects that do not run in families. These genetic effects are the result of complex interactions across loci at the level of the genome and across behaviors at the level of the developing person. Only identical twins inherit all the genes associated with these higher-order interactions, and so they will be much more similar to each other compared to other pairs of genetically related relatives. Some social scientists maintain that high identical twin similarity on temperament measures is due to monozygotic twin assimilation effects (i.e., parents treat identical twins more alike compared to fraternal twins) or to measurement problems (i.e., measures are not sensitive enough to detect moderate to low fraternal twin or sibling similarity). Still, few researchers would argue that temperament is completely determined by the environment.
During the 1990s, the search for biological correlates of temperament differences expanded to include investigations of brain activation patterns. Scientists found brain activation differences between children who approach new situations (i.e., behaviorally uninhibited children) compared to children who withdraw from novel contexts (i.e., behaviorally inhibited children). Even though temperament styles appear to be linked to genetic, physiological, and neurological processes, temperament researchers still consider environmental factors to be very important.