Overlapping Genetic Influences on Executive Functions, General Intelligence, and a General Factor of Psychopathology

Psychiatric disorders are widely comorbid, which can be represented as a latent factor, p, representing a general vulnerability to psychopathology. We used a population-based sample of child and adolescent twins (n = 1,913,... [ view full abstract ]

Psychiatric disorders are widely comorbid, which can be represented as a latent factor, p, representing a general vulnerability to psychopathology. We used a population-based sample of child and adolescent twins (n = 1,913, ages 7.8-20.1 years, M age = 13.1 years, 49% female, 58% non-Hispanic White, 35% monozygotic) from the Texas Twin Project (Harden KP, Tucker-Drob EM, Tackett JL. The Texas twin project. Twin Res. Hum. Genet. 2013 Feb; 16(1):385) to examine the relationship between p and cognitive abilities, specifically, executive functions (EFs) and general intelligence (g). Psychopathology was assessed using child self-report and parent report of 12 symptom domains, including thought problems, somatic problems, neuroticism, and symptoms of externalizing, internalizing, and attention-related DSM-defined disorders. Executive functions (EF) were measured using a battery of tests assessing inhibition, switching, updating, and working memory. Intelligence was measured using the WASI-2, comprising two tests of verbal reasoning and two tests of spatial reasoning. The mean full-scale IQ in the sample was 103 (SD = 14), mirroring population norms. Replicating previous work, a bifactor model with a general p factor fit well. Agreement between child- and parent-report of psychopathology was moderate (r = .4 for p factor). Consistent with previous reports from this sample (Engelhardt LE, et al. Strong genetic overlap between executive functions and intelligence. J. Exp. Psychol.-Gen. 2016 Sep;145(9):1141), latent factors representing general EF and g were highly heritable (h² = 92% for EF, h² = 90% for g) and were nearly synonymous genetically. Across the entire IQ range and across reporters, cognitive abilities were negatively related to p (r ~ -.25), with a stronger association at the low end of ability for parent-reported psychopathology. The associations between cognitive abilities and p were driven by shared genetic variants. Results are interpreted within a cognitive epidemiology framework.