Identifying neural representations of multisensory processing in autism and schizophrenia
Loading...
Date
2024-05-20
Authors
Μιζάρας, Φωκίων
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The main objective of this master thesis is to understand the multisensory integration process in individuals with autism and schizophrenia, by examining the electroencephalogram (EEG signals). More specifically, our aim is to extract meaningful neural representations and patterns from the EEG data, which will be useful for evaluating the behaviour of the participants, and for finding similarities and differences between autistic and schizophrenic people as well as compared to control individuals. The methodology that was followed to produce the results, included statistical analysis at the behavioural level, time-frequency analysis, tensor decomposition and classification at the electrophysiological level.
In the first chapter a detailed review of the current literature, regarding the definition of multisensory integration process and its effects in patients with Autistic Spectrum Disorder and Schizophrenia was performed. In addition, the state of the art in the specific field was highlighted, with emphasis given also to the bibliographic gap, which was later covered in this master thesis.
In the second chapter the methodology utilised in this study was described and explained in detail. Initially, the different populations, participated in the neural experiment, as well as the structure of the experiment were analysed thoroughly. Briefly described, a clinical experiment has been conducted with 90 participants: 23 autistic patients, 35 people with schizophrenia and 32 controls. In this experiment, three different stimuli, audio, visual and audiovisual, were presented to the participants and their neural (EEG) responses were recorded. Moreover, the acquisition and pre-processing of EEG data were mentioned extensively. Furthermore, all the techniques used to produce results at behavioural and electrophysiological level, such as statistical analysis and ANOVA, time-frequency analysis using wavelet transformation, PARAFAC tensor decomposition method and LDA classification algorithm were illustrated and explained in a concise manner.
The third chapter included the results of this master thesis, which were presented and described in detail.
Finally, the fourth chapter included a detailed discussion section, in which all our findings were evaluated and assessed based on the current literature. Additionally, important limitations and future improvements were elucidated for further investigation. Eventually, we reached a conclusion that the SZ population is distinguished from the other two groups, ASD, and CN, in accordance with our behavioural and electrophysiological findings. However, this conclusion could not be generalized to constitute a groundbreaking truth
Description
Keywords
Autism, Schizophrenia, Neurophysiology