Further examination of the relationship between overshoot and filtering

In a previous experiment, threshold was measured for a brief duration probe positioned temporally at the onset of a broadband masker, with and without a broadband precursor, to determine the amount of overshoot. The probe level was then fixed 10 dB above the threshold levels determined in the overshoot experiment, and a notched-noise masker was varied in level to estimate filter shape, with and without a precursor. In general, filters were broader and the signal-to-noise ratio at threshold, k, was lower following a precursor, but there was not a clear relationship between the amount of change in the filters and overshoot. In the present study, filter shapes were measured with the probe level set at the threshold from the overshoot experiment, rather than 10 dB above. Results revealed that the change in k is in general closely related to the amount of overshoot (which would be expected) and also that there is at least some relationship between the change in filter width and the amount of overshoot. The study was also extended to look at possible filtering changes with decrease in overshoot, and to look at effects of overshoot from off-frequency precursors and maskers. [Work supported by NIH (NIDCD).]