Quantitative EEG Signal Analysis in the Study of Stuttering: Estimation of Cortical Sources

Developmental stuttering is a communication disorder that affects speech fluency, characterized by the interruption, repetition and prolongation of words and sounds. It typically develops in early childhood and often resolves spontaneously, although in some cases it persists into adulthood, compromising the social, psychological, and professional lives of individuals who stutter. Consequently, recent years have seen numerous studies employing neuroimaging techniques to investigate the underlying mechanisms of stuttering. These studies have highlighted structural and functional differences in brain regions involved in speech, motor control, and auditory processing. This dissertation aims to study EEG signals to analyse the cortical sources in specific regions of interest in adults who stutter with the goal of providing further information on developmental stuttering. EEG signals were acquired in two resting conditions: with open and closed eyes. The data were then pre-processed and filtered in the frequency range [1 – 45] Hz. Power spectral density (PSD) was calculated for each frequency band (delta [1 – 4 Hz], theta [4.5 – 7.5 Hz], alpha1 [8 – 10 Hz], alpha2 [10.5 – 12 Hz], beta1 [12.5 – 16 Hz], beta2 [16.5 – 21 Hz], beta3 [21.5 – 30 Hz], and gamma [30.5 – 45 Hz]) on both scalp signals and current densities obtained from the neural sources estimation. Additionally, partial directed coherence (PDC) was computed on the regions of interest composing different networks to assess intra – network connectivity. The PSD analysis revealed an overall increase in stuttering in the alpha range in almost every network considered, especially in the open eyes condition and in brain regions of the left hemisphere. Additionally, beta power, which is often linked to the “readiness” of the system to engage in movements, was found to be lower in individuals who stutter with respect to fluent speakers, particularly in sensorimotor regions. Lower delta levels were also observed, especially in frontal regions and in the closed eyes condition. Connectivity analysis revealed abnormal dynamics in the open eyes condition, especially in nodes of the default mode network (DMN), the speech – motor network (SPMN), and in nodes of the cortico – basal ganglia – thalamo – cortical (CBCT) loop. In the closed eyes condition, abnormal modulation of sensorimotor regions was also observed in individuals who stutter, in particular in the SPMN and the social cognitive network (SCN). Although it remains unclear whether these anomalies represent a core deficit of stuttering or arise as a compensatory mechanism, the results of this thesis contribute to further understanding the neural dynamics underlying developmental stuttering in adults and lay the groundwork for future studies aimed at clarifying the nature of these abnormalities and their potential role in therapeutic interventions. These studies have highlighted structural differences in brain regions involved in speech, motor control, and auditory processing. Additional differences are observed when comparing the neural activity in the brains of individuals who stutter with those of neurofluent individuals. This dissertation aims to study EEG signals to analyse the cortical sources in specific regions of interest in adults who stutter with the goal to provide further information on developmental stuttering.