Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Structure–Function Coupling in Adults Who Stutter: Evidence from Multimodal Connectivity Modeling
Poster Session A, Friday, September 12, 11:00 am - 12:30 pm, Field House
Hasini Weerathunge1, Yanni Liu1, Mike Angstadt1, Jason Tourville2, Lauren Keith1, Soo-Eun Chang1; 1Department of Psychiatry, University of Michigan, USA, 2Department of Speech, Language, and Hearing Sciences, Boston University, USA
Introduction: Altered brain structure and function have been independently reported in adults who stutter (AWS), particularly in motor cortical and subcortical circuits, suggesting widespread disruptions in speech-related brain connectivity. However, the relationship between structural and functional connectivity—quantified as structure–function coupling (SFC)—has not been explored in this population. SFC reflects how well structural connections (SC) predict functional interactions (FC), and reduced SFC has been linked to symptom severity in conditions such as ADHD (Hearne et al 2019), Schizophrenia (Jiang et al 2021), traumatic brain injury (Wang et al 2021), and Alzheimer’s disease (Cao et al 2020). These findings suggest that studying SFC may reveal subtle neural abnormalities in stuttering more effectively than unimodal approaches. Prior research shows reduced structural integrity in left hemisphere speech–motor networks and increased functional activity in right hemisphere homologues in AWS (Matsuhashi et al., 2023). This has been interpreted as compensatory right hemisphere recruitment in response to persistent left-hemispheric deficits. Based on this, we hypothesized that AWS would show reduced SFC in left hemisphere speech–motor areas and enhanced SFC in the right hemisphere, reflecting compensatory mechanisms. Methodology: We retrospectively analyzed multimodal MRI data from 65 adults (33 AWS, 26 AWNS), including resting-state fMRI and diffusion-weighted imaging. Functional connectivity was computed using the CONN toolbox (Whitfield-Gabrieli & Nieto-Castanon, 2012) from time series in regions defined by the Speech Lab atlas (Rowe et al 2024). Structural connectivity was estimated using constrained spherical deconvolution implemented in MRtrix3 (Tournier et al., 2019) and quantified as normalized streamline counts between regions. For each participant, we modeled FC as a function of SC using subject-level linear regression, controlling for age, sex, and group. The absolute difference between predicted and observed FC yielded a subject-level error matrix, representing regional SFC decoupling. Group-level comparisons were performed on these error matrices to identify regional differences in SFC. Results: Preliminary analyses from 30 participants (15 AWS, 15 AWNS) revealed no significant group differences in subject-specific SC beta coefficients of the linear regressions (t = -0.937, p = .357), though there was a trend towards reduced SFC in AWS. Regionally, AWS showed greater coupling between left–right subcortical regions (thalamus, caudate, pallidum), but more decoupling between cortical–subcortical connections (e.g., left somatomotor cortex–thalamus, dorsal inferior frontal gyrus–putamen). Moreover, AWS exhibited stronger interhemispheric coupling in motor areas but reduced intra-hemispheric coupling between sensory and motor regions. Conclusion: This study is the first to investigate structure–function coupling in developmental stuttering. Although group-level differences were not significant in this preliminary sample, regional SFC patterns suggest altered brain integration in AWS, particularly involving cortical–subcortical and sensorimotor pathways. Understanding these structure–function dynamics may offer insights into the neurobiology of stuttering and inform future interventions targeting specific neural networks.
Topic Areas: Disorders: Developmental, Speech Motor Control