Adoption studies can be used to test for passive rGE effects in the association between overall family functioning and child antisocial behavior. The adoption method is based on the fact that adopted children and their adoptive parents are not genetically related and, on average, only share a common postnatal environment, therefore the adoptive family environment is free from passive rGE. A stronger association in biological families in comparison to adoptive families is suggestive of passive rGE (Leve et al., 2010). For example, one adoption study found that the association between family functioning and adolescent externalizing problems was greater in biological than adoptive families, confirming the important role of passive rGE (McGue et al., 1996). However, studies examining specifically the role of parenting practices (rather than overall family functioning) indicate that the effects of passive rGE on the association between parenting and child and adolescent antisocial behavior are small to nonexistent (Harold et al., 2013; Marceau et al., 2012; Narusyte et al., 2011). In contrast, their results highlighted the role of evocative (reactive) rGE, the phenomenon explained in the section “Possible Implications of Behavioral Genetic Research.”
The classical adoption method (“adoption at birth”) cannot be used to disentangle prenatal environmental influences (e.g., maternal smoking during pregnancy, maternal depression during pregnancy) from genetic influences, because biological mothers provide both genes and prenatal environment to their children (i.e., passive rGE with prenatal environment). A novel adoption design (“adoption at conception”) allows a separation of the influences of prenatal environment from genetic influences on the child’s development. The design is possible because of modern in vitro fertilization (IVF) technologies that are used to help people with fertility problems to have a baby. During IVF, an egg is removed from the woman’s ovaries and fertilized with sperm in a laboratory. The fertilized egg is then returned to the woman’s womb to grow and develop. During this procedure, an egg and sperm can be used from biological parents, or alternatively, from egg and/or sperm donors. The “adoption at conception” design is based on the comparison of pairs of parents and children with different degrees of genetic relatedness: both parents are genetically related to their child (homologous IVF); only father is genetically related to the child (egg donation); only mother is genetically related to the child (sperm donation); neither of the parents is genetically related to the child (embryo donation) (Rice et al., 2009; Thapar et al., 2007). If a prenatal factor is associated with the child’s trait in both genetically related and genetically unrelated parent–child pairs, then this factor is more likely to be environmental (not influenced by common genetic factors). If a prenatal factor is associated with the child’s trait only in genetically related parent–child pairs, but not in genetically unrelated parent–child pairs, then it is likely to involve passive rGE.
An example of an “adoption at conception” study is the Cardiff IVF study. The study participants were families who had conceived a child through one of the assisted reproductive methods and were subsequently recruited through a number of different fertility clinics that agreed to participate (18 clinics in the United Kingdom and 1 in the United States). In total, 779 families with children born following successful IVF treatment were included. Using data from this study, researchers found that the associations maternal smoking during pregnancy and antisocial behavior of the child (at 4–10 years) were likely to be influenced by genetic factors shared by mothers and children, proving the important role of passive rGE in this association (Rice et al., 2009).
In the case of evocative rGE, an individual’s genetically influenced characteristics evoke particular environmental experiences or responses. For example, children exhibiting externalizing behavior would elicit negative parenting consistent with their genetic propensities towards externalizing behavior (Klahr et al., 2013). One line of evidence for the role of evocative rGE in the etiology of the association between parental negativity and adolescent antisocial behavior comes from the study of more than 700 pairs of siblings, including twins, full siblings, half siblings, and unrelated siblings (Pike et al., 1996). Results of this study showed that the overlap between parenting and antisocial behavior was primarily mediated by genetic factors, suggestive of the important role of evocative rGE. Adoption studies have also found that children (O'Connor et al., 1998) and adolescents (Ge et al., 1996) at genetic risk for antisocial behavior (as indexed by birth mother self‐reported antisocial behavior) were more likely to receive negative parenting from their adoptive parents, confirming the important role of evocative rGE.
In the case of active rGE, a person actively seeks, modifies, or creates an environment that corresponds with their genetic propensities. For example, a study of MZ twins showed that within a pair, the twin with a higher level of antisocial behavior in early adolescence was more likely to have deviant friends in late adolescence (Burt et al., 2009). Similarly, another twin study demonstrated that antisocial behavior in boys was a predictor for selection of deviant peers (the so‐called “social selection”). This study also found the environmental impact of deviant peers on antisocial behavior (the so‐called “social conditioning”). It was shown that social conditioning was more important during childhood, while social selection became important during adolescence (Kendler et al., 2008).
To summarize, there are three different types of gene–environment correlation that can contribute to social development. It is important to note that the effects of different types of rGE may vary with age, with passive rGE being more important in infancy and early childhood, whereas evocative and active rGE play more significant roles in late childhood and adolescence.
Gene‐environment interaction
Gene–environment interaction (G×E) refers to a phenomenon of gene–environment interplay, whereby genetic effects on a trait or disorder depend on the environment, or environmental effects depend on genetic factors. There are different types of G×E interactions. For example, genetic factors can influence a child’s response to adverse experiences. This G×E type is implied in a diathesis‐stress model. Diathesis is a term referring to constitutional predisposition or vulnerability to a particular disease or abnormality. In the diathesis‐stress model, genetic factors serve as vulnerability factors for children who experience one or more adversities (e.g., poverty, maltreatment), so that children with genetic risk are more susceptible to the effects of adversities in the development of psychopathology, whereas children without genetic risk can be more resilient to the development of psychopathology even when adversities are present (Rutter, 2006).
The first evidence for an interaction between a latent (estimated rather than measured) genetic risk and measured environmental adversity in the development of antisocial behavior came from an adult adoption study (Cadoret et al., 1995). The authors demonstrated that having an adverse environment, and a genotypic risk (indexed by having a biological parent showing criminality), increased the risk for antisocial outcomes more than would be expected as just an additive effect of both. Jaffee and colleagues (2005) reported a similar finding based on data from a sample of 5‐year‐old twins. They found that maltreatment was only a significant risk for childhood antisocial behavior in the presence of a high genetic risk, as conferred by having a MZ twin with conduct disorder (Jaffee et al., 2005). Molecular genetic studies have also demonstrated G×E: the effect of childhood maltreatment on antisocial behavior was found to be stronger in people with a less active form of the MAOA gene, and this effect is larger for males than females (Byrd & Manuck, 2014; Kim‐Cohen et al., 2006).
Another G×E type refers to the situation when genetic factors amplify a child’s sensitivity to both positive and negative environments. This G×E type is implicated in a differential susceptibility model: children genetically more susceptible to negative influences (e.g., poverty, maltreatment) could also be more susceptible to positive influences (e.g., educational provision, social support) (Belsky & Pluess, 2009). For example, in a study of the DRD4 gene, maternal sensitivity, and behavioral problems in children, it was shown that low maternal sensitivity was associated