Dept. of Surgery, Univ. of Maryland School of Medicine, 419 W. Redwood St., Baltimore, MD
Johns Hopkins Univ., Baltimore, MD 21218
Univ. of Maryland School of Medicine, Baltimore, MD
Temporal gap detection (TGD) thresholds were measured between sinusoids that varied in frequency from 2000 to 3100 Hz for combinations of 2, 3, or 4 markers (e.g., combinations of one or two pre-gap markers with one or two post-gap markers). Sinusoidal frequencies F[sub 1] and F[sub 4] were used as pre-gap markers, while F[sub 2] and F[sub 3] served as post-gap markers. TGD thresholds were measured from three normal-hearing adults who tracked 70.7% correct detection thresholds adaptively across blocks of 50 2AFC trials. For symmetric marker conditions, where a pair or pairs of pre- and post-gap markers are equivalent in frequency on both sides of the silent gap (e.g., F[sub 1]=F[sub 2] or F[sub 1]=F[sub 2] and F[sub 3]=F[sub 4]), TGD thresholds are small (i.e., <10 ms). However, for asymmetric marker conditions, where only three markers are presented on a trial (e.g., F[sub 1]=F[sub 2], F[sub 2](not equal to)F[sub 3], no F[sub 4]), performance is highly variable. TGD thresholds for asymmetric marker conditions may be very large (e.g., 20--80 ms), and tend to increase as the frequency separation is increased between markers F[sub 3] or F[sub 4] and markers F[sub 1]=F[sub 2]. These patterns of TGD results are inconsistent with a masking process and probably reflect mechanisms that are important in grouping of auditory images. [Research supported by NIH and The Center for Language and Speech Processing, Johns Hopkins University.] [sup a)]Present address: AT&T Bell Laboratories, 200 Laurel Ave., Middletown, NJ 07748.