METHODS
Subjects
Twenty-eight Japanese volunteers (15 males and 13 females, 19–43 years old) participated in the EEG experiments; 12 Japanese volunteers (8 males and 4 females, 19–34 years old) participated in the PET experiment; and 26 Japanese volunteers (15 males and 11 females, 18–31 years old) participated in the psychological experiment. None of the subjects had any history of neurological or psychiatric disorders. Written informed consent was obtained from all subjects before the experiments. The PET and EEG experiments were performed in accordance with the approval of the Committee of Medical Ethics, Graduate School of Medicine, Kyoto University. All subjects were familiar with the actual sounds of the instruments used as a sound source.
Sound materials and presentation systems
Traditional gamelan music of Bali Island, Indonesia, a natural sound source containing the richest amount of high frequencies with a conspicuously fluctuating structure, was chosen as the sound source for all experiments. A traditional gamelan composition, “Gambang Kuta,” played by “Gunung Jati,” an internationally recognized gamelan ensemble from Bali, was recorded using a B&K 4135 microphone, a B&K 2633 microphone preamplifier, and a B&K 2804 power supplier, all manufactured by Brüel and Kjær (Nærum, Denmark). The signals were digitally coded by Y. Yamasaki's high-speed one-bit coding signal processor (United States Patent No. 5351048) (
Yamasaki 1991) with an A/D sampling frequency of 1.92 MHz and stored in a DRU-8 digital data recorder (Yamaha, Hamamatsu, Japan). This system has a generally flat frequency response of over 100 kHz.
Most of the conventional audio systems that have been used to present sound for determining sound quality were found to be unsuitable for this particular study. In the conventional systems, sounds containing HFCs are presented as unfiltered source signals through an all-pass circuit and sounds without HFCs are produced by passing the source signals through a low-pass filter (
Muraoka et al. 1978;
Plenge et al. 1979). Thus the audible low-frequency components (LFCs) are presented through different pathways that may have different transmission characteristics, including frequency response and group delay. In addition, inter-modulation distortion may differentially affect LFCs. Therefore it is difficult to exclude the possibility that any observed differences between the two different sounds, those with and those without HFCs, may result from differences in the audible LFCs rather than from the existence of HFCs. To overcome this problem, we developed a bi-channel sound presentation system that enabled us to present the audible LFCs and the nonaudible HFCs either separately or simultaneously. First, the source signals from the D/A converter of Y. Yamasaki's high-speed, one-bit coding signal processor were divided in two. Then, LFCs and HFCs were produced by passing these signals through programmable low-pass and high-pass filters (FV-661, NF Electronic Instruments, Tokyo, Japan), respectively, with a crossover frequency of 26 or 22 kHz and a cutoff attenuation of 170 or 80 dB/octave, depending on the type of test. Then, LFCs and HFCs were separately amplified with P-800 and P-300L power amplifiers (Accuphase, Yokohama, Japan), respectively, and presented through a speaker system consisting of twin cone-type woofers and a horn-type tweeter for the LFCs and a dome-type super tweeter with a diamond diaphragm for the HFCs. The speaker system was designed by one of the authors (T. Oohashi) and manufactured by Pioneer Co., Ltd. (Tokyo, Japan). This sound reproduction system had a flat frequency response of over 100 kHz. The level of the presented sound pressure was individually adjusted so that each subject felt comfortable; thus the maximum level was approximately 80–90 dB sound pressure level (SPL) at the listening position.
Using the bi-channel sound presentation system, four different sound combinations were prepared as follows:
1) full-range sound (FRS) = HFC + LFC;
2) high-cut sound (HCS) = LFC only;
3) low-cut sound (LCS) = HFC only; and,
4) baseline = no sound except for ambient noise. All experiments were performed in an acoustically shielded room. In the PET experiment, there was a very low-level fan noise from the PET scanner, which did not annoy the subjects. Figure
1 A shows the averaged power spectrum of the source signal obtained from the music with a CF-5220 fast Fourier transform (FFT) analyzer (Ono Sokki, Tokyo, Japan) over an analysis period of 200 s. It contained a significant amount of HFCs above the audible range, often exceeding 50 kHz and, at certain times, 100 kHz. Figure
1 B shows the averaged power spectra of the actual sounds reproduced with a 22 kHz cutoff frequency for the filter and recorded at the subject's head position. The spectrum of FRS was essentially the same as that of the source and contained both LFCs below and HFCs above 22 kHz. None of the blindfolded subjects could distinguish LCS (i.e., HFC only) from silence when it was presented alone. Therefore we concluded that the HFC employed in the present experimental setting was, at least, a consciously unrecognizable air vibration.