Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Atypical Coupling between Speech and Association Networks in Childhood Stuttering
Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai
Jaeeun Kim1, Yanni Liu1, Mike Angstadt1, Soo-Eun Chang1; 1University of Michigan
Background: Developmental stuttering is typically understood as a speech motor disorder, with atypical function and anatomy frequently reported in brain regions supporting speech production and motor control (Neef and Chang, 2024). Recent developmental neuroscience theories suggest that specialized neural circuits, including those supporting speech, do not emerge in isolation but instead develop through interactions with distributed brain regions organized into large-scale networks (Johnson, 2011). For instance, association networks such as the default mode network (DMN) increasingly segregate from sensorimotor networks during typical development (Luo et al., 2024). This network segregation is posited to support specialized and efficient neural processing, whereas disruptions in these patterns have been implicated in various neurodevelopmental conditions (Uddin et al., 2025). We hypothesized that children who stutter (CWS) would show aberrant functional connectivity (FC) between key speech-related regions and distributed association networks, including the DMN and frontoparietal control network (FPCN) and that these alterations would relate to stuttering severity. Methods: Resting-state FC was examined in 70 CWS and 79 controls (3–10 years). Eight association-network regions of interest (ROIs) (four DMN, four FPCN) and four speech ROIs (two somatomotor-lateral, one auditory, and one action-mode networks) were defined using coordinates with high inter-subject network assignment probabilities (>75%) from Dworetsky et al. (2021). Speech ROIs’ relevance was further validated using Neurosynth meta-analytic z-scores. Seed-to-voxel analyses, in which FC values were computed between the speech ROIs and every other voxel, and ROI-to-ROI analyses, in which FC values were computed between all predefined ROIs, were performed. Group differences and associations with stuttering severity measured via SSI-4 were assessed using general linear models controlling for age, sex, IQ, and motion. Network identities of significant seed-to-voxel clusters were inferred based on the probabilistic network profiles of peak voxels. Results: CWS relative to controls showed increased FC (i.e., reduced anticorrelation) between a speech-auditory ROI in the left posterior superior temporal gyrus and a precuneus cluster corresponding to the DMN, suggesting reduced segregation between speech and internally oriented association networks. In contrast, greater stuttering severity was associated with increased FC between bilateral ventral motor cortex speech ROIs and clusters corresponding to the dorsal attention network (DAN) and the FPCN in the left superior frontal and right middle frontal gyri, respectively, linking greater symptom severity to increased coupling between speech-motor and attentional/executive control networks. Conclusions: These findings suggest that atypical interactions between speech-related and association networks are implicated in developmental stuttering. Importantly, distinct association networks showed differential relationships with stuttering, such that increased coupling between speech-auditory and DMN regions emerged as a trait-like group difference, whereas increased coupling between speech-motor and attentional/executive control systems scaled with stuttering severity. Increased speech-auditory–DMN coupling may reflect reduced segregation during development, potentially interfering with the development of auditory feedback mechanisms that are crucial for early speech acquisition. In contrast, increased coupling between speech-motor and top-down control systems may reflect excessive recruitment of executive resources during speech execution in children with more severe stuttering. Together, these findings highlight the role of altered speech–association-network interactions in the neurodevelopmental mechanisms underlying childhood stuttering.
Topic Areas: Disorders: Developmental, Speech Motor Control