Aristotle University Of Thessaloniki (AUTH)
Διερεύνηση αντιληπτικών αποκρίσεων σε μουσικά ερεθίσματα μέσω ανάλυσης ηλεκτροεγκεφαλογραφήματος με προηγμένες τεχνικές επεξεργασίας σήματος
Abstract
dc:descriptionThe aim of this thesis was to investigate two folds of music perception, i.e., the relationship between music and motion and the recognition of music preference, through the analysis of electroencephalogram (EEG) signals. In the first part, the sensorimotor activation of musicians and non-musicians due to audiovisual stimulation concerning orchestral performance-direction and motion-reflecting programmatic music was investigated, through targeted experimental procedures. The level of sensorimotor activation was detected through the quantification of the EEG mu rhythm complexity by the estimation of its fractal dimension, while a relevance analysis of audiovisual stimuli-based characteristics was also implemented using the relevance ratio. Stronger sensorimotor activity was found during the passive listening of orchestral performance and of motion-reflecting music, in the case of musicians, while the relevance of characteristics of the aforementioned stimuli was higher in the case of the trained subjects. The latter outcomes were interpreted in the context of musical education and experience. In the case where visual stimuli relating to the musical excerpts was also present, stronger sensorimotor activity in the case of musicians or a convergence in the response between the latter and non-musicians was observed, by combining results of the fractal and relevance analysis, depending on whether the visual stimulus presented motion that leads to the production of music or not. In the second part of the thesis, the focus was placed on the discrimination of listeners’ electrophysiological (EEG) responses during the listening of liked and disliked musical excerpts, by systematically investigating various parameters concerning both the field of neurophysiology and the fields of signal processing and pattern recognition. EEG data was acquired through an experimental procedure, while the subsequent feature extraction was based on time-frequency distributions of the signals. According to the results, highest classification accuracy (86.52%) was achieved through features extracted from the beta and gamma EEG bands during the whole duration of music listening. Moreover, two time intervals were detected in which classification accuracy reached its peak depending on the level of familiarity with 15 s-long musical excerpts, i.e., 7.5-10.5 s in the case of familiar excerpts (91.02%) and 10-15 s in the case of unfamiliar ones (87.10%). Finally, topological neurophysiological correlates concerning music appraisal responses were also presented based on the EEG features that led to highest classification rates.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2012
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzidimitriou, Stylianos
- Χατζηδημητρίου, Στυλιανός
Subjects
dc:subject × 16- Επεξεργασία σήματος
- Ηλεκτροεγκεφαλογράφημα (ΗΕΓ)
- Μουσική
- Αντίληψη
- Signal processing
- Electroencephalogram (EEG)
- Music
- Perception
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Natural Sciences
- Computer and Information Sciences
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/30784