Level dependence in the two-tone masking experiment

Abstract

The two-tone masking experiment used in the direct estimation of critical bandwidth was conducted at two sound-pressure levels of the masking tones, 50 and 90 dB SPL, on three adult listeners with normal hearing and two with moderate (45 and 50 dB THLs), flat, unilateral, sensorineural hearing losses. The test signal was a narrow band of noise (±80 Hz) centered around 1000 Hz. The two-tone maskers were 1000 Hz±40, ±60, ±80, ±100, ±150, ±200, ±300, and ±500 Hz. In addition to the two-tone masking experiment, control experiments in which each of the single tones of each two-tone pair were used as the masker were conducted at both levels. For listeners with impaired hearing, the experiments were run at both levels in their normal ear but only at the 90-dB level in their impaired ear. For the 90-dB masking level, the two-tone masking experiment failed to show the sudden diminution in masking that defines the critical band at the 50-dB level. This failure to observe the CBW at the 90-dB level can be explained in terms of the control experiments, which indicated that masking by the low tone alone remains relatively constant across frequency for a 90-dB level of the masker.