Correction to: Behavior Genetics (2021) 51:223–236 https://doi.org/10.1007/s10519-021-10044-0

In the original version of this article, Fig. 2 was inadvertently published with a typographic error. In Fig. 2, the covariance matrix between DZ twins should include the components VC_DZ and VE_DZ in the formulae for the variances and not VC_MZ and VE_MZ as originally presented. This typographic error does not affect the results or conclusions of the manuscript. The original version of Fig. 2 has been corrected by publishing this correction article.

The correct Fig. 2 is shown below.

Fig. 2
figure 2

Extension of the Augmented Classical Twin Design to Include Genetic Dominance. Phenotypes for each twin (P1, P2) are modelled as a function of latent additive genetic (A), dominance genetic (D), common environmental (C) and unique environmental (E) sources of variation. The variance of each latent variable is estimated and the path coefficients between the latent and observed variables are constrained to one. The additive genetic covariance between MZ twins is estimated as VA whilst the additive genetic covariance between the ith DZ twin pair is modelled as a function of their estimated genomewide proportion of alleles IBD () multiplied by the additive genetic variance. The dominance genetic covariance between MZ twins is estimated as VD whilst the dominance genetic covariance between the ith DZ twin pair is modelled as a function of the estimated proportion of the genome in which they share both alleles IBD (̂p2,i) multiplied by the dominance genetic variance. Common environmental and unique environmental variables in MZ and DZ twin pairs are allowed to have different variances with the constraint that the total environmental variance in MZ and DZ twins is the same (i.e. VC_MZ + VE_MZ = VC_DZ + VE_DZ). The expected phenotypic covariances under the full model for each zygosity are displayed. Under this formulation, a test of the EEA is whether common and unique environmental variance components

Full size image