THE EFFECTS OF SIGNAL FREQUENCY AND ABSOLUTE BANDWIDTH ON GAP DETECTION IN NOISE

Whether temporal resolution in noisebursts is primarily determined by the highest frequency component in the signal or its absolute bandwidt h remains unclear. In this study, the absolute bandwidths and upper cu toff frequencies of signal noisebursts were varied across broad freque ncy ranges, several times greater than previously jointly studied. The purpose was to determine how each independently affects detection, ta king into consideration that bandwidth effects at one signal frequency might be very different from bandwidth effects at another. Gap detect ion thresholds were obtained for five subjects with normal hearing in a 2 IFC paradigm. Signals were noisebursts whose bandwidths and upper cutoff frequencies varied among 1, 2, 4, 8, and 12 kHz. Their duration was 150 ms and they were presented at an overall level of 75 dB SPL a bove a 45 dB SPL white noise floor. The largest mean gap detection thr eshold, 6.98 ms, was obtained for a noiseburst with a bandwidth of 1 k Hz and upper cutoff frequency of 12 kHz. The smallest mean gap detecti on threshold, 2.22 ms, was found with a bandwidth and upper cutoff fre quency of 12 kHz. Significant interactions were found to exist between absolute bandwidth and upper cutoff frequency. Although gap detection thresholds generally decreased with increasing signal frequency and b andwidth, the pattern was complex. When the absolute bandwidth was at least one-half the upper cutoff frequency then upper cutoff frequency and not bandwidth determined gap sensitivity; but when the absolute ba ndwidth was less than one-half of the upper frequency, then both deter mined gap thresholds. Aspects of these data support the conclusion tha t the high frequency components of signals and the corresponding high frequency auditory fibers are critical to temporal resolution but show , in addition, that acuity is enhanced only in relatively wideband sig nals.