Structural Brain Correlates of Speech Disfluency in Early Childhood: A Dimensional Analysis in a Population-Based Cohort

Abstract

Most neuroimaging studies of speech disfluency have compared individuals who stutter with fluent controls. However, treating speech disfluency as a continuous, dimensional trait offers new insights into the neural basis of fluency during early childhood. We investigated whether naturally occurring variation in speech disfluency is associated with grey matter structure in a population-based sample of 5-year-old children (N = 120; 65 M, 55 F). Disfluency was derived from audio and video-recorded conversational speech samples transcribed and coded using the Ambrose and Yairi classification into stuttering-like disfluencies (SLDs) and other disfluencies (ODs). T1-weighted magnetic resonance imaging data were analyzed using voxel-based morphometry and complemented by surface-based measures of cortical thickness and surface area (FreeSurfer). Whole brain voxel-wise regression tested associations between grey matter and disfluency while adjusting for age, sex, handedness, and maternal education, with additional robustness checks (including intracranial volume adjustment and alternative SLD operationalizations). Higher SLDs were positively correlated with regional grey matter volume primarily in the frontal cortex, with clusters in the left middle frontal gyrus and right superior frontal gyrus showing the greatest cross-model stability (uncorrected voxel-level p < 0.001; FDR-corrected cluster-level p < 0.05); additional regions, including posterior cerebellum, were more sensitive to model specification. No significant voxel-wise associations were observed for ODs, and surface-based analyses yielded no significant associations with SLDs. Boys showed higher SLD rates than girls, but voxel-wise sex × SLD interactions were not significant. These findings suggest that dimensionally measured SLDs across a population-based cohort are most consistently related to frontal control systems implicated in planning and monitoring demands during speech. The overlap with frontal regions reported in diagnosis-based stuttering studies suggests that frontal control circuitry is relevant to fluency variation in early childhood across both dimensional and categorical approaches. These findings support a complementary, dimensional approach to studying fluency variation in early childhood, while underscoring the need for longitudinal work to determine how naturally occurring disfluency relates to later developmental stuttering.