Comodulation masking release for three types of modulator as a function of modulation rate

To investigate the dependence of 'Comodulation Masking Release' (CMR) on the type and frequency of modulator, thresholds were measured for a 4-kHz tone masked by modulated and unmodulated noises. The maskers were a 400-Hz wide band of noise centered on 4 kHz (NOCUE), the same noise with a 2700-Hz lowpass noise added (LPCUE), and a wideband noise with a passband between 3 and 6 kHz (WBCUE). In each condition three types of modulator were used: a square wave, a sinewave, and a lowpass noise. Several frequencies of each type of modulator were used, allowing measurement of the interaction between envelope frequency (fe) and modulator type. Thresholds were lower in the presence than in the absence of modulation for all modulator types, except when fe was high (e.g. 100 Hz). The 'Modulated-Unmodulated Difference' (MUD) decreased monotonically with increasing fe in all conditions. For any given fe, MUDs generally were greater for square-wave and lowpass-noise modulation than for sinusoidal modulation. As significant MUDs were sometimes obtained even in the NOCUE condition, the MUD was not an accurate measure of CMR. Therefore, CMR was defined as the difference between the MUD obtained in the NOCUE condition and that obtained in each cue condition, for each combination of fe and modulator type. Measured this way, the CMR for our 4-kHz signals did not vary systemically with fc over the range studied. When a CMR was obtained it was smaller for sinusoidal modulation than for either square-wave or lowpass-noise modulation.(ABSTRACT TRUNCATED AT 250 WORDS)