Combined polygenic scores boost IQ prediction

Updated: 2026.04.04 1H ago 5 sources
Aggregating GWAS results for intelligence and related traits (notably years of education) produces multipolygenic scores that explain substantially more variation in measured intelligence than single‑trait scores — the review reports combined scores explaining over 10% of variance and accounting for ~20% of the heritable component. This quantitative jump transforms polygenic scores from weak correlates into variables of practical predictive use in longitudinal and policy research. — Greater predictive power makes polygenic intelligence scores relevant to education policy, clinical uses, reproductive decisions, and debates over fairness and privacy.

Sources

Polygenic Score Prediction Within and Between Sibling Pairs for Intelligence, Cognitive Abilities, and Educational Traits From Childhood to Early Adulthood | Published in Intelligence & Cognitive Abilities
2026.04.04 75% relevant
The study reports robust PGS prediction across development and compares cognitive traits to anthropometrics, reinforcing that aggregated/combined PGSs can strongly predict cognitive outcomes in population samples even as within-family prediction is smaller, which connects to claims that scaling GWAS/PGS increases predictive power.
The new genetics of intelligence - PMC
2026.04.04 92% relevant
The article reports that recent large GWAS have identified many variants whose aggregated effects (polygenic scores) account for a substantial share of measured cognitive variance — precisely the empirical basis summarized by the existing idea that combined polygenic scores substantially improve prediction of IQ and related life outcomes.
Genetic variation, brain, and intelligence differences | Molecular Psychiatry
2021.02.02 90% relevant
The article reviews DNA‑based heritability estimates and the discovery of loci from large genome‑wide association studies (GWAS), directly supporting the existing claim that combining GWAS results into polygenic scores improves prediction of cognitive test performance; the authors report modest but growing predictive power and discuss genetic correlations with education and health outcomes.
Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links to intelligence - PubMed
2018.07.04 90% relevant
The paper (Savage et al., 2018) reports new genome‑wide significant loci and functional links from a meta‑analysis of 269,867 people, directly increasing the variant set available to build polygenic scores for cognitive ability and therefore improving predictive power — precisely the mechanism captured by the existing idea about combined polygenic scores boosting IQ prediction.
The new genetics of intelligence | Nature Reviews Genetics
2018.01.08 100% relevant
Key claim in the review: “More than 10% of the variance in intelligence can be predicted by multipolygenic scores derived from GWAS of both intelligence and years of education.”
← Back to All Ideas