A new longitudinal analysis from researchers at King’s College London suggests that childhood intelligence predicts the pace of adolescent brain development. The study, which drew on data from the Adolescent Brain Cognitive Development study, followed 10,495 participants from around nine or ten years of age across four time points into mid‑adolescence.
Participants were tested across three cognitive domains: fluid cognition, crystallized cognition, and matrix reasoning, allowing researchers to see whether different forms of intelligence relate to distinct brain‑development trajectories.
Brain imaging captured three facets of development: structural MRI to measure gray matter volume, surface area and thickness; diffusion imaging to assess white matter microstructure; and resting‑state functional MRI to gauge how brain networks communicate at rest.
Analyses adjusted for household socioeconomic status to account for environmental influences, and for participant sex and scanner differences across sites.
Across the whole brain, baseline fluid cognition predicted how cortical surface area and volume changed over time, with higher initial scores linked to a faster decline in gray matter, particularly in the frontal lobe.
By contrast, cortical thickness showed no significant association with any cognitive measure, and crystallized cognition and matrix reasoning did not relate to long‑term gray matter structure.
White matter development, measured by fractional anisotropy, revealed a different pattern: higher matrix reasoning scores in childhood forecast a slower, more protracted increase in white matter integrity over the study period, starting from a higher baseline FA in several key tracts, including those connecting the frontal lobe with other regions.
Fluid and crystallized cognition did not show a significant link with white matter maturation.
Resting‑state connectivity analyses showed that early cognitive scores predicted trajectories of between‑network communication. Participants with lower baseline scores tended to exhibit increasing connectivity between networks over time, while those with higher scores showed a faster reduction in coupling between networks such as auditory processing and cognitive control.
On a closer scale, higher crystallized cognition and matrix reasoning were associated with a stronger decoupling between the default mode network, active during internal thought, and networks involved in higher‑order attention, with adolescence bringing greater specialization in youths with higher cognitive abilities.
The authors stress that the study is observational and cannot establish causation; shared genetic or environmental factors may influence both intelligence and brain maturation.
Because data from the final measurement point were incomplete, the team relied on linear models, noting that brain maturation is often non‑linear and that future analyses covering a wider age range could map these changes more precisely.
The study, authored by Phoebe Thomson, Philip Shaw and Divyangana Rakesh, is titled Cognitive ability in childhood predicts adolescent structural and functional brain development: A longitudinal study.
