The Gap Detection Test: Can It Be Used to Diagnose Tinnitus?

Abstract

Objectives: Animals with induced tinnitus showed difficulties in detecting silent gaps in sounds, suggesting that the tinnitus percept may be filling the gap. The main purpose of this study was to evaluate the applicability of this approach to detect tinnitus in human patients. The authors first hypothesized that gap detection would be impaired in patients with tinnitus, and second, that gap detection would be more impaired at frequencies close to the tinnitus frequency of the patient.

Design: Twenty-two adults with bilateral tinnitus, 20 age-matched and hearing loss-matched subjects without tinnitus, and 10 young normal-hearing subjects participated in the study. To determine the characteristics of the tinnitus, subjects matched an external sound to their perceived tinnitus in pitch and loudness. To determine the minimum detectable gap, the gap threshold, an adaptive psychoacoustic test was performed three times by each subject. In this gap detection test, four different stimuli, with various frequencies and bandwidths, were presented at three intensity levels each.

Results: Similar to previous reports of gap detection, increasing sensation level yielded shorter gap thresholds for all stimuli in all groups. Interestingly, the tinnitus group did not display elevated gap thresholds in any of the four stimuli. Moreover, visual inspection of the data revealed no relation between gap detection performance and perceived tinnitus pitch.

Conclusions: These findings show that tinnitus in humans has no effect on the ability to detect gaps in auditory stimuli. Thus, the testing procedure in its present form is not suitable for clinical detection of tinnitus in humans.

Fig. 1.

Mean audiograms for the Tin group, the NoTin group, and the Con group. The error bars indicate the group standard deviations around the mean. Con indicates young normal-hearing subjects; NoTin, matched group without tinnitus; Tin, tinnitus.

Fig. 2.

Tinnitus loudness ratings before and after the gap detection test, shown for each Tin subject and each session. The ratings were measured by means of a numeric rating scale from 0 (inaudible) to 10 (tinnitus sounds as loud as imaginable).

Fig. 3.

Mean gap detection thresholds of the four test stimuli at three stimulus levels (5, 10, and 25 dB SL) for the Tin group, NoTin group, and Con group. The gap detection thresholds are averaged across sessions. The error bars indicate the group standard deviations around the mean. Con indicates young normal-hearing subjects; NoTin, matched group without tinnitus; Tin, tinnitus.

Fig. 4.

Gap detection thresholds related to perceived tinnitus pitch. Subjects were stratified with respect to the perceived pitch of their tinnitus. A, “<4.0 kHz.” B, “4.0–5.0 kHz.” C, “5.0–6.3 kHz.” D, “6.3–8.0 kHz.” E, “>8.0 kHz.” F, “WB.” In each of these six sets, the gap detection thresholds of 19, 5, 4, 16, 16, and 5 subjects, respectively, were included. Red symbols reflect cases where the frequency band of the stimulus used to assess the gap detection threshold overlapped the perceived tinnitus pitch. The error bars indicate the standard deviations around the mean.