Department   Undergraduate School  , School of Science and Technology
   Position   Professor
Language Japanese
Publication Date 2017/02
Type Academic Journal
Peer Review Peer reviewed
Title Detection of music preferences in young adults and elderly indi-viduals using prefrontal hemodynamics
Contribution Type Sole-authored
Journal Transactions of the Japanese Society for Medical and Biological Engineering
Journal TypeJapan
Publisher Japanese Society for Medical and Biological Engineering
Total page number null
Details Partaking in physical exercise while listening to one's favorite music ameliorates the perception of dyspnea and fatigue and increases one's enjoyment of the exercise. Hence, the use of specific music in a physical training class for elderly people could increase the endurance of the participants. Additionally, this could benefit elderly individuals by reducing frailty and improving cognitive function. However, it may be difficult to elicit information pertaining to preferred music as it may be difficult to communicate with elderly individuals owing to cognitive decline. Thus, in this study, the objective classification method of music preferences was proposed using the change in prefrontal hemodynamic signals while listening to music. This was measured by functional near-infrared spectroscopy (fNIRS). The experiment consisted of 17 young adults and 17 elderly individuals who listened to 6 popular songs that were released during different time periods. During the activity, fNIRS was used to scan the prefrontal brain activity of the subjects. While the test subjects were listening to each song, seventeen features of the fNIRS waveform were extracted, which included statistical measures of the temporal signal distribution and their laterality with oxy-hemoglobin, deoxy-hemoglobin, and total hemoglobin concentration changes. From these, three features that exhibited the highest correlation to the subjective preference scale obtained from questionnaires completed by each participant were selected. The extracted features were used to train a 2-class linear classifier that determined whether the listener preferred each song. Mean classification accuracy was cal-culated by a leave-one-out cross-validation method. The proposed algorithm indicated a mean classification accuracy of 86.3 +/-11.8% and 88.9 +/- 14.5% (mean +/- standard deviation) in detecting individual favorite songs in young adults and elderly individuals, respectively. The mean classification accuracy was significantly higher in the case wherein the features were individually selected as compared to the case in which the fixed and common features were used with respect to the participants. This suggested that the individual changes in emotional responses were evoked when listening to their favorite songs.