The Etiology of Social and Nonsocial Components of Autistic Behavior in Young Twins

Background: Autistic symptoms are present in individuals without a diagnosable autism spectrum disorder and show quantitative variation in the general population (Hoekstra et al., 2007).  Further, it has been shown that in a general population sample of school-age children, the genetic influences on autistic-like social behaviors are only modestly overlapping with genetic influences on nonsocial symptoms, with most of the factors being specific to either one class of behaviors or the other (Ronald et al., 2005, 2006). 

Objectives: The purpose of this study was to explore the etiology of the social and nonsocial components of ASD symptoms in a younger sample of twins, because it is possible that the genetic architecture underlying these behaviors differs across age groups.  The age of two is particularly salient because it is frequently acknowledged that this is the age when autism can first be reliably diagnosed (DiCicco-Bloom et al. 2006).

Methods: Participants included 312 same-sex pairs of 2-year-old twins (144 MZ, 168 DZ).  Autistic-like behaviors were assessed via parent ratings on the pervasive developmental problems subscale of the CBCL (Achenbach & Rescorla, 2000).  This subscale was divided into behaviors that were social in nature (e.g., poor eye-contact and peer relationships) and those that were considered nonsocial (i.e., relating to restrictive/repetitive behaviors and interests).  Cholesky bivariate decomposition models were fitted to the data.

Results: Intraclass correlations were stronger for identical than for fraternal twins for both social and nonsocial autistic-like behaviors, suggesting that both categories of behaviors were genetically influenced to some extent.  Several Cholesky decomposition models were fitted to the data.  The full ACE model accounts for genetic factors, as well as shared- and nonshared environmental influences; reduced versions of this model were also fit, with an AE (genetics and nonshared environment) being the best-fitting model.  This reduced model showed both social and nonsocial ASD behaviors to be influenced by genetic factors (.52 and .47, respectively), but with limited overlap between the two categories of behavior, with only 13% of the variance in nonsocial behaviors explained by common genetic influences from social behaviors.  In the full model, which also includes shared environmental influences, this common genetic variance component drops to 5%, with shared environment accounting for an additional 11%. 

Conclusions: While both social and nonsocial autistic-like behaviors are highly heritable in 2-year-olds, genetic model-fitting analyses show that distinct genetic factors are influencing each cluster of symptoms with little overlap.  These findings are consistent with previous findings in school-age children (Ronald et al., 2005, 2006).  The results suggest that social and nonsocial symptoms are inherited to some extent independently; considering them separately might aid future molecular genetics research.