Association between the Oxytocin Receptor Gene (OXTR) and Childhood Aggression

Human aggression has important implications for long-term life outcomes and is a complex trait, influenced by a multitude of factors. Behavior genetic studies have found that aggression is highly heritable across the lifespan.... [ view full abstract ]

Human aggression has important implications for long-term life outcomes and is a complex trait, influenced by a multitude of factors. Behavior genetic studies have found that aggression is highly heritable across the lifespan. Despite its high heritability, molecular genetic studies of aggression have had mixed results, with few genetic variants showing reliable associations. Recent research suggests that Oxytocin and the Oxytocin Receptor Gene (OXTR) influences social cognition and behavior in humans and animal models. The primary aim of this study is to examine associations between OXTR and measures of aggression in children and adolescents. Based on current animal models, it is hypothesized that OXTR will be associated with various forms of aggression. In the current investigation we collected DNA and parent ratings of aggression for a total of 636 children ages 6-18 years old sampled from both unreferred twins as well as clinically-referred children. The investigation included 31 SNPs in OXTR. To operationalize OXTR in a gene-based test, a series of Exploratory Structural Equation Models (ESEMs) of the OXTR SNPs were conducted. The model with five factors fit best while still favoring parsimony and was used to characterize the underlying structure of the gene. Aggression was also modeled as both a unitary aggression factor and as two separate reactive and proactive aggression factors. A series of gene-based tests were then conducted such that the aggression factors were regressed on the latent OXTR factors and all covariates. The OXTR factors accounted for a significant portion of the variance in the unitary aggression factor (R² = 1.8%, p<.001), reactive aggression (R² = 1.2%, p<.001), and proactive aggression (R² = 5.8%, p<.001) over and above the contribution of the covariates. Future research will focus on replicating these results in larger samples from a diverse group of cohorts.