Index of content:
Volume 108, Issue 2, August 2000
- PHYSIOLOGICAL ACOUSTICS 
108(2000); http://dx.doi.org/10.1121/1.429599View Description Hide Description
It is often assumed that at frequencies in the tuning-curve tail there is a passive, constant coupling of basilar-membrane motion to inner hair cell (IHC) stereocilia. This paper shows changes in the phase of auditory-nerve-fiber (ANF) responses to tail-frequency tones and calls into question whether basilar-membrane-to-IHC coupling is constant. In cat ANFs with characteristic frequencies efferent effects on the phase of ANF responses to tail-frequency tones were measured. Efferent stimulation caused substantial changes in ANF phase (range −80° to +60°, average −15°, a phase lag) with the largest changes at sound levels near threshold and 3–4 octaves below characteristic frequency (CF). At these tail frequencies, efferent stimulation had much less effect on the phase of the cochlear microphonic (CM) than on ANF phase. Thus, since CM is synchronous with basilar-membrane motion for low-frequency stimuli in the cochlear base, the efferent-induced change in ANF phase is unlikely to be due entirely to a change in basilar-membrane phase. At tail frequencies, ANF phase changed with sound level (often by 90°–180°) and the from a fiber was positively correlated with the slope of its phase-versus-sound-level function at the same frequency, as if were caused by a 2–4 dB increase in sound level. This correlation suggests that the processes that produce the change in ANF phase with sound level at tail frequencies are also involved in producing It is hypothesized that both efferent stimulation and increases in sound level produce similar phase changes because they both produce a similar mix of cochlear vibrational modes.
A high-precision magnetoencephalographic study of human auditory steady-state responses to amplitude-modulated tones108(2000); http://dx.doi.org/10.1121/1.429600View Description Hide Description
The cerebral magnetic field of the auditory steady-state response (SSR) to sinusoidal amplitude-modulated (SAM) tones was recorded in healthy humans. The waveforms of underlying cortical source activity were calculated at multiples of the modulation frequency using the method of source space projection, which improved the signal-to-noise ratio (SNR) by a factor of 2 to 4. Since the complex amplitudes of the cortical source activity were independent of the sensor position in relation to the subject’s head, a comparison of the results across experimental sessions was possible. The effect of modulation frequency on the amplitude and phase of the SSR was investigated at 30 different values between 10 and 98 Hz. At modulation frequencies between 10 and 20 Hz the SNR of harmonics near 40 Hz were predominant over the fundamental SSR. Above 30 Hz the SSR showed an almost sinusoidal waveform with an amplitude maximum at 40 Hz. The amplitude decreased with increasing modulation frequency but was significantly different from the magnetoencephalographic (MEG) background activity up to 98 Hz. Phase response at the fundamental and first harmonic decreased monotonically with increasing modulation frequency. The group delay (apparent latency) showed peaks of 72 ms at 20 Hz, 48 ms at 40 Hz, and 26 ms at 80 Hz. The effects of stimulus intensity, modulation depth, and carrier frequency on amplitude and phase of the SSR were also investigated. The SSR amplitude decreased linearly when stimulus intensity or the modulation depth were decreased in logarithmic steps. SSR amplitude decreased by a factor of 3 when carrier frequency increased from 250 to 4000 Hz. From the phase characteristics, time delays were found in the range of 0 to 6 ms for stimulus intensity, modulation depth, and carrier frequency, which were maximal at low frequencies, low intensities, or maximal modulation depth.