This study examined the modifying role of accuracy of informant assessments for heritability estimates of personality and temperamental traits. Data from 7199 individuals including data of 1118 twin pairs were analyzed to estimate heritability on the basis of twin correlations of peer assessments for different levels of familiarity between peers and target twins treated as an index of information accuracy of peer assessments. Moreover, interactions between estimates of genetic effects and information accuracy were tested for self-report-specific variance not accounted for by peer reports. Heritability estimates on the basis of correlations between peer reports on twins increased with the degree of peer reports’ accuracy, whereas it decreased for self-report residuals due to an increase of the genetic overlap between self- and peer-reports with the degree of information accuracy. This indicates (1) that estimates of heritability of temperamental and personality traits are larger when modeling reports of informants who know the targets well and (2) that genetic variance in self-reports on personality traits indeed reflects accurate trait variance rather than genetic variance in self-rater biases.
Read full abstract