Speech recognition in noise and temporal resolution abilities of formally trained musicians

Abstract

Background: It has often been reported that musicians may have an advantage, or a ‘musician advantage’, to perceive speech in the presence of background noise. Musicians may also have improved central auditory processing skills, which may assist in better speech recognition in noise, however contradicting research has found no musician advantage in speech recognition in noise. Aim: The aim of the current study was to compare the temporal resolution and speech recognition in noise abilities between musicians and non-musicians with normal hearing sensitivity (PTA of  15 dB HL). Method: In this quantitative cross-sectional comparative group research study forty participants’ conventional and extended high-frequency pure tone thresholds were determined using the hearTest™. The Digits-in-Noise (DIN) test was used to evaluate speech recognition in two different listening conditions, namely homophasic and antiphasic. Temporal resolution was tested using the Gaps-in-Noise (GIN) test. A binaural intelligibility-level difference (BILD) was calculated by using the two values obtained in the homophasic and antiphasic recognition in noise listening conditions. Results: There were no differences between the groups in EHF thresholds, DIN SNR values and BILD SNR values, as well as the GIN approximate thresholds. Musicians presented with slightly better performances, however not significant (p > .05), in the GIN and DIN tests. Conclusion: This study could not find significant differences in the specified auditory performances between musicians and non-musicians. The musician group, however, performed slightly better, although not significantly different, across all tests, except with the BILD results. The non-musician group performed better, although not significantly better, in the BILD test. The results of this study suggest longitudinal research in order to shed more light on discrepancies in literature. Keywords: musicians; temporal resolution; speech recognition in noise; extended high frequencies