Abstract
Music has a significant influence on human emotions; however, understanding its underlying neural mechanisms remains a complex challenge. This study investigates music-induced emotional responses using synthetically generated electroencephalography (EEG) signals. A controlled synthetic EEG dataset is developed to simulate brain activity across diverse demographic groups exposed to self-selected musical preferences. The generated signals undergo preprocessing, feature extraction, and classification using machine learning and deep learning models. Experimental results indicate that synthetic EEG signals can capture distinguishable emotional patterns, achieving a classification accuracy of 58.55% under multi-class conditions. Although the study does not replace real EEG analysis, it provides a scalable and reproducible framework for controlled experimentation. The findings highlight the potential of synthetic data in affective computing and brain–computer interface research, particularly in scenarios where real EEG acquisition is constrained.