Musical therapy is used for the functional training for dementia, developmental disorder, depression, and so on. The therapy reliefs the stress of the subject and causes the relaxation effect. The effect depends on the music preferences. We have studied the electroencephalogram (EEG) which relates to the preference. In the last meeting, it was preliminarily reported that the prefrontal theta and the left prefrontal alpha bands may participate in the preference. It has not yet been known whether EEG is kept for the preferences, and the common characteristics of EEG exists or not.
Four healthy males (24.5 ± 0.5 years (mean ± s.e.m.)) participated in the experiment. In the experiment, they listened to the music for 30 sec, thought for 15 sec, and then decided the preference at four levels. After that, there was a rest for 20 sec. A subject listened to the thirty pieces in 10 genres. EEG was recorded throughout the experiment. It was analyzed using fast Fourier transform. The mean power levels of theta (4-8 Hz), alpha (8-13 Hz), beta (13-30 Hz), and gamma (30-50 Hz) were calculated while the subject listened to the music, and while they chose the preference. In addition, cross-validation using the power levels from Fast Fourier Transformation was performed with the liner discriminant analysis. The additional experiments were performed a week (day2), a month (day3), and six months (day4) after the first experiment.
The music preferences at each subject gradually different with time. Even when the subject listened to the same groups of music and have different preferences, recorded EEG was also different. In each experiment, EEG which related to the music preferences in each subject was different. On the other hand, the common features of EEG which related to the music preferences were found. They were the right frontal gamma, the right central beta and gamma, the left parietal gamma. Using these features, the discrimination accuracy was about 68 % at day1. The accuracy increased once at day2, gradually decreased, and at day 4 six month after the first experiment, the accuracy was above 60 %.
These results suggest that there is the common feature of EEG reflecting the music preference, and the EEG is redundant for at least six months.