Detection of unexpected tones in gated and continuous maskers.

Abstract

The effect of gating a wideband masker on the detectability of tones having unexpected frequencies was assessed using a probe-signal method. This method leads the listener to expect a target frequency by presenting the signal most often at that frequency, and measures sensitivity to other unexpected frequencies via occasionally presented probe tones. For 295-ms signals, the probe-signal contours (percent correct as a function of probe frequency) of two of four subjects were considerably broader for a 295-ms masker than for a continuous masker. For 5-ms signals, the probe-signal contours of four of five subjects were quite broad and similar for both gated and continuous maskers. When the probe-signal contours were expressed as the attenuation in decibels of the probes, the resulting "probe-signal filters" were frequently broader than auditory filters measured using notched noise in the same subjects. This suggests that subjects may monitor multiple auditory filters under some conditions in the probe-signal task. Signal threshold tended to be higher for conditions showing wider probe-signal filters, indicating a potential link between changes in signal threshold due to masker gating and the number of frequency channels that are monitored.