ASEBA Research

Debates about whether people are shaped more by nature or nurture are giving way to research on how particular aspects of nature and nurture may interact. A study by Jaffee et al. (2005) clearly demonstrates how genetic factors and environmental factors may interact to promote conduct problems. As part of the E-Risk (Environmental Risk) Longitudinal Twin Study, 1,116 pairs of 5-year-old British twins were assessed via home interviews with mothers and with the CBCL and TRF. Conduct problems were measured by summing scores from the Aggressive Behavior and Delinquent Behavior (now called Rule-Breaking Behavior) syndromes, plus DSM-IV conduct disorder (CD) and oppositional defiant disorder (ODD) symptoms. The empirically based CBCL and TRF data were then combined with the diagnostically based DSM-IV data. The correlation between mother and teacher reports of conduct problems was .29, similar to the mean correlation between parent and teacher reports found in meta-analyses of cross-informant correlations. Genetic risk for conduct problems was estimated for each twin from the co-twin's CD status and zygosity. To assess a potential environmental contributor to conduct problems, Jaffee et al. used extensive interviews with mothers to identify children who were physically maltreated. When conduct problems were measured by summing CBCL/TRF syndrome scores and DSM-IV CD and ODD symptoms, genetic factors accounted for 72%% of the variance. On the other hand, when CD was diagnosed on the basis of reports of >3 DSM-IV CD symptoms, genetic factors accounted for 58% of the variation in diagnoses. Reports of maltreatment were also significantly associated with conduct problem scores and diagnoses. Furthermore, there were significant statistical interactions between genetic risks for conduct problems and reports of maltreatment. These interactions reflected the fact that, as levels of genetic risk increased, maltreated children scored progressively higher on conduct problems than children not reported to be maltreated. In other words, maltreatment had bigger negative impacts on children who were at high genetic risk than on children who were at lower genetic risk. A similar pattern was found for CD diagnoses, although the associations were weaker, owing to the lower statistical power afforded by diagnoses than by the quantitative measures of conduct problems. The authors concluded that their "findings showed that such early-onset conduct problems were most likely to emerge when genetically vulnerable children were maltreated. The Genetic Risk x Maltreatment interaction was associated with a clear and significant elevation in conduct problem symptoms."