DTA

Digital Theses Archive

 

Tesi etd-03222023-155619

Type of thesis
Dottorato
Author
COMETA, ANDREA
E-mail address
andrea_cometa@live.it
URN
etd-03222023-155619
Title
Advanced statistical frameworks to uncover the electrophysiological correlates of syntactic processing
Scientific disciplinary sector
ING-INF/06
Course
Istituto di Biorobotica - PHD IN BIOROBOTICA
Committee
Relatore Prof. MICERA, SILVESTRO
Membro Prof. RICCIARDI, EMILIANO
Membro Prof.ssa ZOLLO, LOREDANA
Keywords
  • neurolinguistics
  • stereo-eeg
  • eeg
  • syntax
  • event-related causality
  • intracortical
  • syntactic processing
Exam session start date
24/10/2023;
Availability
parziale
Abstract
This thesis aims at advancing the knowledge of the neural representation of syntactic structures. Unravelling the neural correlates of syntactic processing is particularly appealing for several fields of study, such as linguistics, psychology, neuroscience, and computer science. Moreover, the study of the neural basis of syntax is fundamental in developing new strategies for the treatment of language disorders, such as speech prostheses.<br>The work presented in this thesis focuses on the electrophysiological investigation of syntactic processing, using both non-invasive recordings (electroencephalography - EEG) and intracortical recordings (stereo-electroencephalography - SEEG). These types of recordings offer unique possibilities in the investigation of cognitive functions, such as language processing.<br>The first work presented in this thesis explores the relationship that exists between clinical neuroscience and neurotechnology, with a focus on the positive feedback loop that emerges from their interaction. In the second work, a new statistical framework for the event-related causality estimation in SEEG data is presented. In the third work, this statistical framework is leveraged to investigate the neural networks responsible for the processing of noun phrases and verb phrases. In the fourth work, a novel set of acoustic stimuli is presented and characterized through the use of different notions of surprisal, showing that only the syntactic surprisal can account for the expected differences in the neural responses to the different classes of stimuli. Finally, in the fifth work, EEG recordings of healthy subjects listening to the new stimuli show that the predictability of the syntactic class modulates the response to noun phrases and verb phrases and that the observed brain response is exclusively related to syntactic processing.<br>Overall, this thesis exhibits the possibilities offered by the electrophysiological recordings for the investigation of the neural basis of syntax. It is shown that syntactic processing is a complex task that requires the collaboration between multiple cortical and subcortical areas and that the more complex the syntactic operation, the more brain areas are involved in its processing. The findings presented here are fundamental for the development of strategies for speech impairment mitigation.
Files