Overview
- On Thursday, September 3, 2026, researchers led by Michigan State University published a Nature meta-analysis that pooled data from 46 cohorts and up to roughly 1.14 million people to study the genetics of the Big Five.
- The team reported roughly 1,200 independent genetic signals associated with at least one Big Five trait, with some summaries listing 1,257 signals and others ~1,194 because of different presentation choices.
- Common variants captured in the study explain a modest share of personality differences — about 4.8–9.3% in population analyses and 10.5–16.2% after adjusting for measurement error — so genetics predict group trends but not individual destinies.
- Findings held across regions, age groups, self- and informant-reports and within-family tests of about 51,000 relatives, which argues against simple confounding by shared family environment; Mendelian-randomization gave cautious hints of directional links to health and life outcomes but did not prove causation.
- Polygenic scores trained on the largely European-ancestry data performed substantially worse in African-ancestry groups, underscoring a need for larger, more diverse samples if genetic findings are to be useful across global populations.