An ACE in the hole: Twin family models for applied behavioral genetics research

Abstract

Many studies investigate the genetic and environmental influences on traits using twin data and ACE models (A=additive genetic effects, C=shared environment effects, E=unshared environment effects). Unfortunately, relying on twins leads to biased results and limits what researchers can study. We introduce twin models and describe their problems. We show how to solve these problems with data from twins and their families, which allow modeling new effects such as the parent-to-child transmission of traits. We illustrate twin family models using extraversion (extroversion) data from the Virginia 30,000 twin family dataset, giving model specifications and code for the program Mplus. We conclude that if researchers are interested in understanding a broad sense of genetic influences on observed variables, traditional twin models are adequate. However, when data from twins and families are available, twin family models offer better and more interesting estimates of genetic and environmental effects.