The impact of language background and listening-in-noise on the phonemic restoration effect

Abstract

Listeners often need to understand speech in the presence of background noise or other degraded conditions; doing so is particularly difficult for bilinguals. Phonemic restoration (PR) may be a helpful speech mechanism in noisy settings: this auditory illusion describes the phenomenon where the brain "fills in" missing information, causing listeners to perceive a speech signal interrupted with noise as if it were complete. PR is driven by a combination of top-down and bottom-up factors, and there exists natural variation between groups and individuals in the size of the PR effect that is derived from combining these sources of information. The purpose of this project was to test how listeners' ability to use PR in understanding sentences is affected by number of known languages (monolingual or bilingual), as well as other individual differences between listeners such as the ability to understand sentences in continuous background noise, working memory, and vocabulary knowledge. Phonemic restoration was tested in a broad sample of English monolinguals and bilinguals; neither group showed an overall effect of PR. English monolinguals who were better at comprehending sentences in steady-state background noise did exhibit a PR benefit, though bilinguals with equivalent speech-in-noise-comprehension did not. The results suggest that though individuals vary greatly, monolinguals and bilinguals may use PR differently in understanding sentences, providing further evidence that the effect of bilingualism on listening-in-noise is driven by the use of top-down language-specific information. The finding that only a subset of monolinguals showed PR highlights the need for diverse participant recruitment.