Deficits in reciprocal social behavior in male twins: Evidence for a genetically independent domain of psychopathology

John N. Constantino, James J. Hudziak, Richard D. Todd

Research output: Contribution to journalArticlepeer-review

121 Scopus citations


Objective: Previous studies have demonstrated substantial genetic influences on many child psychiatric disorders, including autism. In this study the authors attempted to quantify the degree to which genetic influences on deficits in reciprocal social behavior (a defining feature of pervasive developmental disorders) are shared with genetic influences on other domains of behavior in children. Method: Child Behavior Checklists (CBCL) and Social Responsiveness Scales (SRS) were completed for an epidemiological sample of 219 pairs of male twins. The SRS (formerly known as the Social Reciprocity Scale) is a measure of social impairment that distinguishes children with autism spectrum disorders from those with other child psychiatric disorders. Results: Regression analysis indicated that CBCL syndromes account for 43% of the variance in SRS scores. Bivariate analyses revealed that SRS scores are affected, in part, by phenotypic influences from the CBCL Social Problem syndrome. Forty-four percent of the causal influences on SRS scores, however, are independent from those on CBCL syndromes and are genetic in nature (90% confidence interval: 0.38-0.49). Conclusion: These results support the existence of a continuous distribution of deficits in reciprocal social behavior in the population, which are substantially genetically independent from other domains of child psychopathology.

Original languageEnglish
Pages (from-to)458-467
Number of pages10
JournalJournal of the American Academy of Child and Adolescent Psychiatry
Issue number4
StatePublished - Apr 2003


  • Autism
  • Genetic epidemiology
  • Pervasive developmental disorders
  • Social responsiveness scale
  • Twins


Dive into the research topics of 'Deficits in reciprocal social behavior in male twins: Evidence for a genetically independent domain of psychopathology'. Together they form a unique fingerprint.

Cite this