Understanding and Preventing Violence, Volume 2: Biobehavioral Influences (1994)
Chapter: STUDIES OF HUMANS
Twin studies, studies of adoptive parent-offspring pairs, and studies of the biologic parents of adoptees are required to untangle the joint effects of genetic and family environmental transmission. Although such designs are becoming routine in the study of cognitive development, they are rare in the study of violent behavior.
The third challenge aims at determining whether a genetic propensity to violence is substantial or trivial and the extent to which it is correlated with other behaviors. Does the genetic influence on intelligence and on alcohol abuse explain genetic liability toward aggression?
STUDIES OF HUMANS
Studies of the inheritance of violent behavior in humans rely on adoption or twin designs to tease apart the effects of shared family environment from those of shared genes. The adoption design capitalizes on the fact that an adoptee does not receive environmental transmission from a biological parent or genetic transmission from an adoptive parent. Similarly, adoptive siblings share environments but not genes, whereas biological siblings raised apart share genes but not environments. Twin studies rely on the fact that identical twins have all genes in common, whereas fraternal pairs share on average only half their genes (plus a small effect from assortative mating). In both kinds of studies, heritability coefficients—the proportion of observed variation due to genetic variation—may be calculated. As with all human behaviors, the interpretation of these coefficients may be confounded by several factors—selective placement in adoptions, and imitation and co-offending in twins. In addition, the study of violence presents problems of its own. Not only is the base rate for violence low, but it is also more poorly measured than most behaviors studies by behavior geneticists.
Despite methodologic weaknesses in the early twin studies, later twin and adoption research suggests important heritability for adult antisocial behavior with perhaps a smaller genetic influence on juvenile criminality (Bohman et al., 1982; Christiansen, 1977; Cloninger and Gottesman, 1987; Mednick et al., 1984). Heritability estimates range from a low of about .20 to a high of almost .70 in Danish samples. Needless to say, such estimates cannot be easily extrapolated to other cultures.
In contrast to these data, the evidence for a genetic basis to violent offending is much weaker. Only three samples permit one
