Environmental Influences on Positive Affect Different Measures, Different Results

Environmental Influences on Positive Affect: Different Measures, Different Results Introduction Results • Positive affect (PA) is one of the few temperamental dimensions that shows moderate to high shared environmental influences in behavioral genetic studies (Goldsmith et al., 1997; Goldsmith et al., 1999, Volbrecht et al., 2007). • Previous quantitative behavioral genetic research of PA, however, has exclusively employed parent-rating measures, which are subjective and may be prone to rater biases. Phenotypic Correlation • Observed PA was modestly correlated with the parentrated PA (r=.17, p< .01). Table 1 Twin Intraclass Correlations and Cross-Twin Cross-Trait Correlations Figure 1. Lab Assessment: Dinky toys Twin zygosity Variables • Moreover, research on other temperamental domains such as activity level has indicated that different measures of the same trait can tap different genetic and environmental effects (Saudino, 2009). Consequently, measure-specific effects may also apply to other temperamental traits such as PA. Figure 2. Lab Assessment: Snack Delay Goals • In contrast, shared environmental effects accounted for over half of the variance in parent-rated PA and genetic and nonshared environmental effects were more modest. MZ twins DZ twins Observed PA .35* .13* Parent-rated PA .69* .62* Cross Correlations .16* .00 Note. MZ=monozygotic; DZ=dizygotic. *p<.05. Twin Intraclass and Cross Correlations (Table 1) • For both observed and parent-rated PA, MZ intraclass correlations exceeded DZ correlations, suggesting that both were genetically influenced. 2)To what extent do behavioral observations and parent ratings of PA tap the same genetic and environmental factors? • For parent-rated PA, DZ intraclass correlation exceeded one-half the MZ correlation, suggesting that shared environmental effects also influenced parent-rated PA. Figure 3. Lab Assessment: Gift Model-Fitting Analyses • 304 same-sex (140 monozygotic, 164 dizygotic) twin pairs assessed within approximately 2 weeks of their 3rd birthday. Measures Observational Measure of PA. • An observational measure of PA was obtained by coding the frequency of smiles/laughter during three episodes (i.e., “Dinky toys”, “Snack delay”, and “Gift”) from the Laboratory Temperament Assessment Battery (LabTAB; Goldsmith et al., 2005). • Smiling/laughter were rated as either 0 (behavior absent) or 1 (behavior present) for a total of 23 behavioral segments. • A bivariate Cholesky model (Figure 4) was fit to the data. The model includes 3 latent factors (A1, C1, and E1) influencing both observed and parent-rated PA , and 3 factors (A2, C2, and E2) accounting for residual influences specific to parent-rated PA. • Using this model we are able to estimate 1) the heritability, and shared and nonshared environmental variance of each domain; 2) genetic and environmental contributions to the phenotypic correlation; and 3) genetic and environmental correlations between observed and parent-rated PA . A1 C1 E1 A2 C2 E2 Model-Fitting Results • Figure 5 presents the path estimates from the best-fitting model. Variance components are presented in Table 2. • Individual differences in observed PA were attributed to moderate genetic and high nonshared environmental factors, but not shared environmental factors. A1 E1 .59 (.45, .68) Parent Rating Measure of PA. • This subscale consists of 10 questions regarding the child's frequency of smiling, laughing, or squealing with joy in specific situations (e.g., playing with favorite toys). • MZ twin cross correlation (i.e., the correlation between Twin A's score for parent-rated PA with Twin B's score for observed PA and vice versa) exceeded DZ twin cross correlation, suggesting genetic influences on the association between observed and parent-rated PA. .76 (.70, .81) • During these episodes, each twin was assessed without his/her co-twin present. • The Pleasure subscale of the Toddler Behavior Assessment Questionnaire (TBAQ; Goldsmith, 1996) provided a parent-rating measure of PA. c2 e2 Observed PA .34 (.20, .47) -- .66 (.53, .80) Parent-rated PA .12 (.04, .23) .58 (.49,.66) .30 (.23, .38) rg rc re 1.00 (1.00, 1.00) -- -- Covariance 1) Would a more objective measure of PA (i.e., behavioral observations) also show shared environmental effects? Sample Table 2 Estimates of Genetic and Environmental Variance (95% CI) from the Best Fitting Model Variance • The present study aimed to address two questions: Methods • The genetic correlation indicated a complete overlap between genetic factors influencing the two measures. It was these overlapping genetic effects that contributed to the phenotypic correlation between the measures. There was no significant overlap in the environmental influences operating on both measures. E2 C2 -.81 (-.89, -.73) -.55 (-.61, -.48) .34 (.20, .48) Observed PA Parent-rated PA Figure 4. Full bivariate Cholesky model. A= genetic factors; C= shared environmental factors; E= nonshared environmental factors. Observed PA Parent-rated PA Figure 5. Path estimates (95% CI) from the best-fitting bivariate Cholesky model. Observed PA— Parent-rated PA h2 Note. h2= genetic variance; c2= shared environmental variance; e2= nonshared environmental variance. rg, rc and re denote the genetic, shared environmental, and nonshared environmental correlations, respectively.