Temporal resolution and CMR can depend upon frequency

Abstract

An estimate of temporal resolution can be obtained by subtracting the signal threshold obtained in the presence of a modulated masker from that obtained in the presence of an unmodulated masker (yielding the modulated‐unmodulated difference, or MUD). An estimate of comodulation masking release (CMR) can be obtained by measuring the MUD with both a broadband and critical‐band masker. The present study obtained MUD and CMR for signal frequencies from 250 to 4000 Hz. The masker was unmodulated or sinusoidally amplitude modulated at rates from 2 to 16 Hz. For all rates and both masker bandwidths, the MUD increased monotonically with signal frequency; it was larger for the broadband masker, indicating an across‐channel CMR. The CMR increased from a few dB at 250 Hz to 14 dB at 4000 Hz. The frequency effect (for MUD and CMR) was considerably reduced when the modulation depth (m) of the masker was decreased from 1.0 to 0.9, and was essentially eliminated at a depth of 0.5. The results of a forward‐masking experiment using the unmodulated masker at both masker bandwidths suggest that the frequency effect is due to differences in the recovery of forward masking, and that suppression can influence CMR (and MUD) at very large modulation depths. [Work supported by NIDCD.]