Performance in detecting a 50 ms pure-tone signal in a longer tone-bur
st masker was evaluated as a function of the delay between masker and
signal onset. Delays from 500 ms to 5 s were used, and detection was m
easured in terms of sensitivity d'. Performance was also measured for
a continuous masker. In a first experiment, the masker and the signal
had the same frequency (500, 1000, 2000, 4000 and 8000 Hz). In this ca
se, sensitivity increased with delay for high frequencies and for high
masker levels. In a second experiment, the masker and the signal had
different frequencies; the masker was set at 4 kHz and the signal freq
uency was inside and outside the critical band of the masker. For this
second case, the results depended on masker level and signal frequenc
y: increasing the delay for a 50 dB SL masker did not improve detectio
n; at 80 dB SPL, however, delaying the signal improved detection for s
ignal frequencies close to the masker frequency. The assumption is mad
e that the improved detection for long durations of the masker origina
tes from a long-term decrease in the firing rate of the auditory nerve
fibers.