Twin studies to GWAS: there and back again

The field of human behavioral genetics has come full circle. It began by using twin/family studies to estimate the relative importance of genetic and environmental influences. As large-scale genotyping became cost-effective, genome-wide association studies (GWASs) yielded insights about the nature of genetic influences and new methods that use GWAS data to estimate heritability and genetic correlations invigorated the field. Yet these newer GWAS methods have not replaced twin/family studies. In this review, we discuss the strengths and weaknesses of the two approaches with respect to characterizing genetic and environmental influences, measurement of behavioral phenotypes, and evaluation of causal models, with a particular focus on cognitive neuroscience. This discussion highlights how twin/family studies and GWAS complement and mutually reinforce one another. Traditional twin/family studies and genome-wide association studies (GWAS) are complementary methods for behavioral genetic research.Twin/family studies estimate aggregate genetic and environmental influences on a trait, while GWAS suggest more specific plausible biological mechanisms.Estimates of genetic (co)variance from these two approaches have different assumptions, require different sample sizes for adequate power, and have distinct interpretations.Twin/family studies typically examine deep phenotypes, such as executive functioning or memory; GWAS tend to focus on general phenotypes (e.g., intelligence), often minimally assessed, that can be administered in large samples or harmonized across multiple studies.Data from both approaches can be used to test causal models. Family data can be used to control for confounds in GWAS and probe gene–environment interplay.