Reactive Inhibitory Control Precedes Stuttering Events

Abstract

AbstractResearch points to neurofunctional differences underlying fluent speech production in stutterers compared to non-stutterers. Considerably less work has focused on processes that underlie stuttered vs. fluent speech within stutterers, partly due to difficulties associated with eliciting stuttering during neuroimaging. We used magnetoencephalography to test the hypothesis that a reactive inhibitory control response in the action-stopping network precedes stuttering events. We leveraged a novel clinical interview to develop participant-specific stimuli (words identified as likely to be stuttered) known to elicit a comparable amount of stuttered and fluent trials across participants. Thirty-one adult stutterers participated (8 female). Participants completed a delayed-response task which included a cue (prior to the go cue) signaling the imminent requirement to produce the word. In response to this cue, stuttered (vs. fluent) productions resulted in greater beta power in the right presupplementary motor area (R-preSMA), a key node in the action-stopping network, a signature of reactive motor inhibition. Beta power was related to stuttering severity and was predictive of stuttering on a trial-to-trial basis. Stuttered words were associated with delayed speech initiation, also consistent with the reactive inhibition hypothesis. Post-hoc analyses revealed that words associated with stronger anticipation (independently-generated vs. researcher-assisted words words) were associated with more stuttering and greater beta power. This points to a critical relationship between reactive inhibition and stuttering anticipation such that words identified more readily by participants as likely to be stuttered elicit greater inhibition. This is the largest and most balanced investigation of stuttered and fluent speech to date. It advances understanding of the neural bases of stuttering events, and offers a target region for future studies that test the efficacy of neuromodulation on stuttering.Significance StatementThe neural bases of stuttered speech are unknown, largely due to difficulties in eliciting stuttered speech reliably in the lab. We used a novel method to elicit stuttered speech during MEG recordings in adult stutterers to test the hypothesis that reactive inhibition precedes stuttered speech. We observed a neural response characteristic of reactive motor inhibition (increased beta power in the right presupplementary motor area) in response to a cue signaling the imminent requirement to produce a word likely to be stuttered. Speech initiation was also delayed for stuttered trials. Beta power was related to stuttering severity and percentage of trials stuttered, and predicted whether individual trials would be stuttered. These results contribute to a brain-based account of stuttering events and have important clinical implications, e.g., providing a target region for future neuromodulation studies.