Twin Differentiation of Cognitive Ability Through Phenotype to Environment Transmission: The Louisville Twin Study

Abstract

The Louisville Twin Study is one of the most intensive twin studies of cognitive ability. The repeated measurements of the twins are ideal for testing developmental twin models that allow for the accumulation of gene-environment correlation via a (P⇒E) transmission process to explain twins' divergence in mean ability level over time. Using full-scale IQ scores from 566 pairs of twins (MZ = 278; DZ = 288), we tested whether a P⇒E transmission model provided better representation of actual developmental processes than a genetic simplex model. We also addressed whether the induced gene-environment correlation alters the meaning of the latent nonshared environmental factors with a simple numerical method for interpreting nonshared environmental factors in the context of P⇒E transmission. The results suggest that a P⇒E model provided better fit to twins' FSIQ data than a genetic simplex model and the meaning of the nonshared environment was preserved in the context of P⇒E.

Keywords: Cognitive development; Genetic simplex; Intelligence; Louisville Twin Study; Nonshared environment.

Figure 1

Theoretical description of “sibling drift.” +CA = initial genetic advantage for cognitive ability relative to co-twin; -CA = initial genetic disadvantage for cognitive ability relative to co-twin.

Figure 2

Generic ML-SEM genetic simplex model. Biometric components of P_it, phenotypic CES-D scores for twin i at time t, are estimated between- and within-families; $A_t^b$ = between-family genetic effect at time t; $E_t^b$ = between-family (common) environmental effect at time t; $A_t^w$ = within-family genetic effect at time t; $E_t^w$ = within-family (nonshared) environmental effect at time t; $u{A}_t^b$ = unique between-family genetic effect at time t; $u{E}_t^b$ = unique between-family environmental effect at time t; $u{A}_t^w$ = unique within-family genetic effect at time t; $u{E}_t^w$ = unique within-family environmental effect at time t; a_ar, c_ar, and e_ar = auto-regression coefficient between adjacent components; the between-family and within-family genetic loadings for the MZ twins are 1 and 0, respectively, to meet the assumption that MZ twins share 100% of their genes; the between-family and within-family genetic loadings for the DZ twins are both √5 to meet the assumption that DZ twins share 50%, on average, of their segregating genes.

Figure 3

Generic ML-SEM genetic simplex model with P=>E. The red line represents that the P=>E parameter, b_PE, which was only estimated at the within-family level in the DZ group.

Figure 4

MZ and DZ intraclass correlations (ICCs) of FSIQ scores from age 4 to age 15.

Figure 5

Stability of within-family differences in DZ and MZ twins’ FSIQ scores. At each wave, t, twins were rank ordered from low to high FSIQ scores within their families. Means at each wave ≥ t were calculated for “low” FSIQ and “high” FSIQ twins, and then plotted over time. The procedure was repeated until all t were exhausted.