Spectral properties of cortical activity associated with speech tasks in Parkinson’s disease

Parkinson’s disease is a neurodegenerative illness characterized by a variety of symptoms, including speech impairments. This thesis addresses in particular the role of the interplay between cortical and subcortical activity in determining these symptoms. This work explores the joint spectral modulation of cortical electroencephalographic (EEG) and subcortical Local Field Potentials (LFP) recordings from Parkinsonian patients acquired both during reading and resting conditions. Furthermore, a connectivity analysis was performed to highlight connections between subcortical LFP and EEG signals in different parts of the cortex. EEG shows desynchronization in the alpha (5-11 Hz) and beta (11-30 Hz) bands, i.e., a decrease in Power Spectral Density (PSD) in Read task compared to the resting condition, primarily involving frontal and parietal channels. LFP exhibited desynchronization in alpha and beta3 (21.5-30 Hz) bands in both left and right subthalamic nucleus (STN), while the left STN showed an increase in PSD for the [11-21.5] Hz frequency range. The most relevant finding of this work concerns the EEG-LFP connectivity: EEG tends to lead the LFP in the alpha and beta bands, particularly during resting condition, whereas this relationship is less clear in the gamma band. Moreover, significant differences in the connectivity pattern were observed between frontal-central cortex and left STN associated with the reading task in beta bands. Finally, the Imaginary part of the Coherency method enabled a functional link with brain areas involved in speech production, coherently with the State Feedback Control model of speech production. Overall, the thesis provides a new characterization of the interplay between cortical and subcortical activity during speech in Parkinsonian patients, paving the way to a deeper understanding of speech impairment symptoms and the possible development of novel therapies.