HomeScienceStatistical Learning Declines Across C
SCIENCE

Statistical Learning Declines Across Childhood, Study Finds

A longitudinal study tracking children from ages 7 to 14 found that statistical learning abilities drop as executive functions develop.

WHAT YOU NEED TO KNOW
  • Researchers tracked statistical learning in the same participants across ages 7, 8, 11, and 14.
  • Data showed statistical learning is most effective in early childhood and declines as executive functions develop.
  • The paper was accepted on 14 August 2026 and published on 3 September 2026 in Nature Communications.

Statistical learning abilities decrease as children grow older, according to a longitudinal study published in Nature Communications. The decline coincides with the growth of executive functions, the cognitive processes that direct goal-oriented behaviour.

Humans rely on statistical learning to detect environmental patterns and anticipate future events, underpinning motor, cognitive, and language capabilities. Prior cross-sectional studies suggested that statistical learning peaked between ages 11 and 13 before dropping in adulthood, but those earlier studies did not track the same individuals over time.

Researchers addressed that limitation by assessing the same cohort across four stages: ages 7, 8, 11, and 14. The authors used linear mixed models and latent class analyses to measure changes across the testing points. The data revealed that statistical learning was most effective during early childhood and progressively declined as children matured and acquired stronger executive functions.

Eszter Tóth-Fáber, Bence C. Farkas, Tímea Tánczos, Dezső Németh, and Karolina Janacsek contributed equally as lead authors. Their team spanned ELTE Eötvös Loránd University and the HUN-REN Research Centre for Natural Sciences in Budapest, the École Normale Supérieure in Paris, the University of Szeged, Université Claude Bernard Lyon 1, Atlántico Medio University in Spain, and the University of Greenwich in the United Kingdom.

Funding came from the French National Research Agency, the National Brain Research Programme, and Hungary's Ministry of Innovation and Technology via the National Research, Development and Innovation Fund. The journal received the submission on 19 August 2024, accepted it on 14 August 2026, and published the peer-reviewed findings on 3 September 2026.

Xentir Media
Xentir Media NewsroomSource-backed AI and technology coverage, drafted by Xentir's automated editorial system under fixed human-set rules. See our editorial policy and AI usage policy.
J
Jomon · Founder & EditorFounder and editor of Xentir Media. Sets the editorial rules the newsroom system runs under, and is accountable for its corrections. About Jomon · [email protected]
The Xentir Brief
The developments worth knowing — one useful email.
Get the Brief →