Monozygotic and dizygotic twins reared together: sensitivity of heritability estimates.

Abstract

This paper examines the method of estimating the heritable component of behavioural traits by comparing observed similarities between pairs of MZ (monozygotic) twins as opposed to DZ (dizygotic) twins. The technique of sensitivity analysis is used to investigate the importance of assumptions made to calculate heritability estimates. This permits not only the investigation of specific assumptions, but also the effects of different combinations of assumptions. The assumptions examined include: the level of assortative mating, degree of intra‐pair environmental similarity of MZ as opposed to DZ twins, and the presence of genotype‐environment covariance. It is noted that estimates of genetic variance—although large—are sensitive to small changes in these assumptions, and that different combinations of assumptions can produce similar estimates of genetic variance. Although this work is effected within the classical twin method, its implications are also valid for more sophisticated analyses of the ‘nature‐nurture’ problem.