Masking release at low sensation levels.

Abstract

Previous results have shown that listeners with sensorineural hearing loss (SNHL) obtain about half of the masking release of their normal‐hearing (NH) counterparts. When speech is amplified sufficiently, listeners with SNHL may score like NH listeners in quiet and in steady noise, yet may obtain only half of the expected release from gated noise. We hypothesize that some of that deficiency may occur because of the impaired listeners’ low speech sensation levels, which results in decreased usefulness of the speech signal in the noise dips. In the current study, NH listeners were tested for their recognition of IEEE sentences in quiet, in steady noise, and in gated noise with the speech presented at varying sensation levels. At low levels (10–15 dB SL), NH listeners scored nearly 100% correct in quiet. In steady noise (at −10 dB signal‐to‐noise ratio) scores for low‐level stimuli were also similar to those obtained at higher SLs. However, at low SLs in gated noise, NH listeners demonstrated less masking release than expected, suggesting that audibility of speech in the dips of noise is important for masking release even when performance in quiet and in steady noise seems optimized. [Work supported by NIH 5R01DC008306.]