2.4. Physiological and Psychological Evaluation

To evaluate ANS activity, we analyzed HRV. We conducted frequency analyses of electrocardiogram waveforms using a heart rate monitor (LRR-03, GMS Co., Ltd., Tokyo, Japan). We used MemCalc analysis system (Tarawa, GMS Co., Ltd.) for real-time spectral analysis with the maximum entropy method. This software calculates heart rate (HR) spectral power from pulse trains composed of the R-R intervals of four heartbeats. The software can separately analyze very low-frequency (LF) components (below 0.04 Hz), LF components (0.04–0.15 Hz), and high-frequency (HF) components (0.15–0.40 Hz). The sampling frequency was set at 250 Hz.

Because HF components also respond to activity in the respiratory center, caution is required when measuring HRV [15]. To control for respiratory condition, we discontinued the experiment if a participant had a maximum respiratory interval shorter than 7 seconds, respiratory frequency below nine breaths/min, or breathing rate greater than 24 breaths/min.

In this study, we analyzed the following indices of HRV. HR is an index of both SNS activity and PNS activity. Increased HR indicates increased SNS activity, while decreased HR indicates increased PNS activity. For PNS activity, we used normalized value HFnu, calculated by dividing the sum of HF and LF by HF. We used the LF/HF ratio as an index for SNS activity. The sum total of all autonomic nervous system activity was denoted by the normalized value LFnu, calculated by dividing the sum of HF and LF by LF. The stress recovery ratio, which was a normalized index of stress recovery, was defined as the ratio of change for each HRV index. This was calculated by subtracting SN from WN and dividing the HRV index observation values by the value obtained by subtracting SN from MS.

In addition to the HRV measurements, subjective evaluations of the emotional impact of the music stimuli were determined using a modified SD method [16, 17]. Fourteen pairs of adjectives were assessed on seven-point scales, including “loud–soft,” “beautiful–ugly,” “pure–impure,” “hard–soft,” “sharp–dull,” “strong–weak,” “deep–metallic,” “annoying–not annoying,” “mild–gruff,” “pleasant–unpleasant,” “powerful–unsatisfactory,” “pleasing–unpleasing,” “shrill–calm,” and “noisy–quiet” [18, 19]. The SD rating test was performed after listening to MS.