목적: People with hearing loss complain of listening to music, and it is even more challenging to perceive the emotions of music. Although a large body of studies are focused on the perception of musical emotions in people with hearing loss, the underlying cortical mechanisms for the perception of musical emotions in people with hearing loss are unclear. In this study, to investigate the effect of audibility, we measured cortical activity response to the emotional perception of music. 방법:Normal hearing group (NHO) and simulated high- (NHH), and low- (NHL) frequency filtered groups were created by using original stimuli and applying low-, and high-pass filtering (1000 Hz cutoff) to musical stimuli, respectively. A total of 48 healthy participants were randomly assigned to three groups (16 people / group). Fifteen musical stimuli developed in our lab were used for the study. The pre- evaluated stimuli were composed of five melodies, and each melody was expressed differently according to emotions including happiness, sadness, and neutrality. During 64-channel EEG recording, participants listened to the randomly presented stimuli binaurally via two speakers followed by ratings of arousal and valence (dimensional model) and selecting emotions (discrete model, thus named discrete). A total of 300 trials were conducted 20 times repeatedly for 15 stimuli. 결과:The NHL group had lower ratings for arousal than the NHH group and for valence than the NHH and NHO groups. To examine the effect of frequency-dependent filtered music, we performed a time-frequency analysis comparing three groups. We also applied surface Laplacian spatial filtering to reduce volume conductivity before time-frequency analysis. We selected trials according to the rating criteria of arousal and valence (happy>7, sad<4, and 4<=neutral<=6) and the percent corrects of perceived emotions. As a result, the topography of the NHL group showed higher alpha activity in parietal channels than the other two groups for arousal ratings of sad, while the NHH group showed higher alpha activity in parietal channels than other groups for valence ratings of sad. 결론:The composer of these stimuli expressed sadness with a slow tempo. This might make it easy for all participants to distinguish sad stimuli. Nevertheless, alpha activity shows that high-frequency audibility which was simulated by low-frequency filtered music affects the arousal while low-frequency audibility which was simulated by high-frequency filtered music affects the valence.