Abstract:

This experiment tested the idea that the auditory system contains a ``modulation filter bank.'' It was an analog in the modulation-frequency domain of an experiment in the audio-frequency domain, whose results were explained in terms of ``ringing'' in the auditory filter [M. J. Shailer and B. C. J. Moore, J. Acoust. Soc. Am. 81, 1110--1117 (1987)]. Subjects had to detect a gap in sinusoidal modulation (rates of 5, 10, 20, 40, and 100 Hz, modulation index =1) applied to a 4-kHz 70-dB sinusoidal carrier. The modulator preceding the gap ended with a positive-going zero crossing. The modulator was turned on at the end of the gap either at a positive-going zero crossing (standard phase) or at a negative-going zero crossing (reversed phase). For very small gaps [<0.1 times the period (P) of the modulator], detectability was very poor for the standard-phase condition but was good for the reversed-phase condition. For low modulation rates, and for a gap duration of 0.5P, the opposite pattern was observed. The results are consistent with what would be expected from ``ringing'' in a modulation filter bank. [Work supported by The Wellcome Trust and the MRC, UK.]