A large international study published in Nature Human Behaviour has found that common DNA variations are linked to toddler temperament in the first three years of life, with genetic patterns that overlap with adult personality, attention‑deficit/hyperactivity disorder (ADHD) and autism. The research was led by Anja Hollowell of Birkbeck, University of London, and senior author Angelica Ronald of the University of Surrey, and drew on data from several European cohorts.
Temperament describes early‑developing personality traits that shape how a child reacts to their surroundings. The study focuses on four main traits: emotionality, activity, shyness and sociability. Emotionality measures how intensely a child experiences negative feelings; activity captures movement energy; shyness reflects inhibition around strangers; and sociability indicates the desire to interact with others.
Observational work over the years has suggested that these early styles can forecast later behaviour. For example, a 2018 infant study co‑authored by Mark H. Johnson linked high activity and poor self‑control in toddlerhood to later ADHD symptoms, illustrating how early temperament can echo into childhood and beyond.
Previous genetic research estimated that 30% to 60% of variation in these temperament traits is heritable, yet could not pinpoint specific DNA differences. The new study aimed to map common DNA variants associated with toddler temperament and test how they relate to later outcomes, preregistered protocols alongside data from five European cohorts: the Norwegian Mother, Father and Child Cohort Study, the Avon Longitudinal Study of Parents and Children in England, the Twins Early Development Study in England and Wales, Lifelines, and the Netherlands Twin Register.
Parents or caregivers assessed emotionality, activity, shyness and sociability at about 18–24 months and again at 36–38 months, with sample sizes ranging from roughly 44,000 to 73,000 children of European ancestry. The research team conducted genome‑wide association analyses adjusted for age, sex and relatedness, scanning millions of common single‑letter DNA variations for links to each trait.
The researchers found that common genetic variation accounts for about 15% of cross‑age differences in shyness, 10% in activity, 7% in emotionality, and 3% in sociability. The genetic signals were stable between the second and third years, with similar results when analyses were run separately for boys and girls.
Across all analyses, ten independent genomic regions met the genome‑wide significance threshold for emotionality, activity or shyness, while no region reached significance for sociability. In aggregate, about 20 temperament‑associated genes were identified, nearly all linked to activity. When these genes were compared with a Genomics England autism gene database, temperament‑linked genes appeared enriched for autism genes at more than ten times the base rate, though this conclusion rests on only two genes.
To explore how these regions might influence the brain, the researchers used colocalization analysis to test whether the same DNA variants also affect gene expression in the cortex, using post‑mortem adult brain tissue. Two regions showed a high probability of sharing an underlying variant with brain expression.
A variant on chromosome 12 near the gene RHEBL1, linked to lower cross‑age activity and higher expression of RHEBL1 in the cortex, carried a roughly 93% probability of a shared underlying variant. This variant has previously been associated with more years of schooling and higher cognitive test scores. RHEBL1 plays a role in cellular signalling pathways involved in brain development and connectivity.
A second region, on chromosome 1 near MR1, showed a 99% probability of shared underlying variant with increased cortical expression of MR1, a gene not previously connected to behaviour. MR1 is involved in presenting microbial byproducts to immune cells, a finding the authors note aligns with evidence linking the immune system to brain development and behaviour. The brain tissue analysed came from adults, and the authors emphasise that laboratory work is needed to understand how these genes influence temperament in children.
Researchers then estimated genome‑wide genetic correlations to see how toddler temperament overlaps with other traits. They found a strong negative relationship between shyness and sociability, as well as moderate negative overlap between emotionality and sociability and between activity and shyness. There was a modest positive link between emotionality and shyness, while correlations between activity and emotionality or sociability were not statistically significant.
When comparing the temperament traits with 17 other traits, from walking age to adult personality, the study reported widespread genetic overlap with neurodevelopmental, psychiatric, cognitive and personality outcomes. Higher activity in toddlers was linked with a moderately strong association with ADHD, earlier walking and a moderate negative association with adult educational attainment and cognitive performance. Higher sociability associated with a lower risk of autism and higher adult extraversion, while higher shyness correlated with lower extraversion, a lower ADHD risk, and slightly later walking. Higher emotionality showed modest positive overlap with autism, adult neuroticism, depression and schizophrenia, with no significant links to lifetime anxiety, bipolar disorder, conscientiousness or agreeableness.
In additional modelling, 86% to 97% of the common variants tied to toddler shyness overlapped with variants linked to ADHD, extraversion and age at walking, though many of these shared variants acted in opposing directions for the different traits.
Polygenic scores, which sum across many DNA variants to predict a trait, explained 0.06% to 1.9% of temperament differences in cohorts not used to discover the signals. Within a fraternal twin framework, differences between within‑ and between‑family genetic predictions were not statistically significant, suggesting limited influence from shared family environments or measurement noise.
The team extended their analysis to non‑European populations by adding Japanese‑ancestry children from Japan’s Tohoku Medical Megabank Birth and Three‑Generation Cohort Study and Black African and mixed‑ancestry children from South Africa’s Drakenstein Child Health Study. In a combined sample of 56,083 to 78,894 children, the chromosome 12 RHEBL1 signal remained linked to third‑year activity, and multi‑ancestry polygenic scores performed similarly to European‑trained scores when tested in European children.
The findings echo earlier work suggesting genetic variation can shape early social‑emotional temperament, such as a seven‑month‑old infant study linking a CD38 gene genotype to cuddliness and touch‑seeking, though the current research examines genome‑wide associations in toddlers rather than a single gene.
Genetic links identified for toddler temperament and brain development
Several caveats accompany the study. Common variants explained only a minority of the variation in temperament (3% to 15%), and polygenic scores accounted for less than 2% of differences, highlighting the probabilistic nature of genetic influence on personality trajectories. The reported regions were not replicated in an independent sample, and cross‑age analyses were not preregistered. The gene‑expression analyses relied on adult brain tissue, not toddler samples, and laboratory follow‑ups are needed to confirm functional roles.
Moreover, the heritability estimates based on common variants were lower than the 30% to 60% suggested by twin studies, with sociability potentially affected by environmental or measurement factors such as family size or questionnaire design. The study notes that temperament was assessed by parent report, which can be influenced by parental perceptions, and that twin comparisons can exaggerate differences in fraternal pairs.
More than 90% of participants in the study were of European ancestry, and fewer than 6,500 children from Japanese and South African cohorts used different questionnaires. The authors call for larger, multi‑ancestry samples, laboratory studies of genes such as RHEBL1 and MR1, and research into how early environments interact with these genetic associations.
The study, titled Genome‑wide association studies of infant and toddler temperament in European and multi‑ancestry populations, lists Anja Hollowell and Anna Gui as joint senior contributors, with a wide group of co‑authors from institutions across Europe. The work adds to the growing evidence that common genetic variation is linked to early behavioural patterns and that these links intersect with later cognitive and psychiatric outcomes.
