Age-Related Differences in Resting-State Functional Connectivity Predict Specific Patterns of Speech Disfluency

Fluent speech production remains largely preserved across adulthood, yet subtle disruptions such as filled pauses, repetitions, and revisions—become more common with age. These disfluencies may reflect underlying cognitive and neural changes that accompany aging, particularly in executive function (EF) and large-scale brain network organization. In this study, we examined whether EF and resting-state functional connectivity (RSFC) independently or jointly explained age-related differences in naturalistic speech disfluencies in an adult lifespan sample (n= 272, ages 20-81 years). We utilized RSFC to assess intrinsic patterns of brain organization, specifically, the degree of within-network, between-network, and segregation measures across three large-scale systems implicated in language and cognitive control: the Language network, the Default Mode Network (DMN), and the Multiple Demand (MD) network. By capturing basal i.e., task-independent connectivity patterns, RSFC provides insight into how the neural architecture of functional networks impacts speech production and its vulnerabilities in aging. Our findings indicate that age was associated with increased disfluency rates, especially for filled pauses and revisions, as well as with lower EF, lower DMN segregation, and reduced connectivity between the Language and DMN networks. EF performance predicted filled pauses, while distinct RSFC patterns predicted revisions and unfilled pauses. Mediation analyses confirmed that EF partially mediated the relationship between age and filled pauses. Additionally, both DMN segregation and Language-to-DMN connectivity partially mediated the relationship between age and revision disfluencies, suggesting that changes in functional brain organization contribute to age-related increases in self-monitoring disruptions. These findings demonstrate that EF and RSFC account for dissociable mechanisms underlying age-related speech disfluency. While EF may support fluent planning, the structure and segregation of large-scale brain networks appear critical for maintaining internal monitoring during speech. This work highlights the utility of disaggregating disfluency types and utilizing cognitive and neural markers to better understand how age-related changes in spontaneous language production.