EVIDENCE FOR AN ACROSS-FREQUENCY, BETWEEN-CHANNEL PROCESS IN ASYMPTOTIC MONAURAL TEMPORAL GAP DETECTION

Monaurally measured temporal gap detection (TGD) thresholds characteri stically increase as the frequency difference is increased over a rang e of about half an octave to an octave between two sinusoids that mark the onset and offset of the silent gap. For greater sinusoidal freque ncy separations, the TGD thresholds often become asymptotic. This patt ern probably reflects two different processes. The first process Likel y reflects within-channel processing within a single auditory filter o r channel. The second process is less certain, but may reflect between -channel processing of the silent gap stimulus across two or more inde pendent frequency channels. To evaluate the hypothesis that asymptotic monaural gap detection can be explained by a simple between-channel p rocess, TGD thresholds were measured as a function of frequency separa tion between a pregap sinusoid presented to the left ear (channel 1) a nd a postgap sinusoid, of higher frequency, presented to the right ear (channel 2). The rationale for dichotic presentation of the sinusoida l markers and gap signal followed from the fact that the gap detection task must be performed between two independent channels by combining the outputs from each channel (ear) and recovering the gap information centrally. The resulting TGD thresholds for pregap sinusoids from 250 to 4000 Hz were relatively invariant and increased only slightly with increasing marker frequency separation. The average TGD thresholds fo r four listeners were in the range of 30 to 30 ms, which corresponded closely with their asymptotic TGD thresholds for the same set of stimu lus conditions measured monaurally. This correspondence of the two dat a sets supports an across-frequency, between-channel process for asymp totic monaural gap detection at marker frequency separations greater t han about half an octave. (C) 1998 Acoustical Society of America.