Communication between native and non-native speakers of English in noise

Abstract

Non-fluency has a negative impact on speech understanding in noise, particularly when hearing protection devices are worn. In multi-national military operations where the communication language is English, it is important to understand the effects of non-native speech and accent on speech understanding. Twenty-four normal-hearing participants were divided into two groups: monolingual English speaking from birth (NA group), and those who learned English after the age of 10 (NN group). All participants completed the Language Experience and Proficiency Questionnaire (LEAP-Q; Marian et al., 2007) to confirm their group assignment. Two experimental sessions were completed, in which each participant was paired with an NA participant in one session and an NN participant in the other. The modified rhyme test (MRT) and speech perception in noise test (SPIN) were administered with each participant pair using two methods. In the first, participants spoke to each other using a communication headset (radio) in background noise of 80 dBA. In the second, the particpants wore the headset with the radio off and spoke to each other face-to-face in background noise levels of 55, 60 and 65 dBA. Performance was calculated as the percentage of correct responses. For the MRT, there was a main effect of talker for both the face-to-face (NA- 79.6%; NN-75.2%), and radio conditions (NA-87.2%; NN- 77.5%). There was also a main effect of background noise level for the face-to-face condition (81.1%, 78.3% and 72.8% for the lowest to highest noise levels, respectively). For the SPIN, there was a main effect of the listener in both the face-to-face (NA-70.9%; NN-54.1%) and radio conditions (NA-86.4%; NN-73.8%), as well as a main effect of background noise level for the face-to-face condition (68.4%, 63.5% and 55.7%). Overall, the results indicate that both NA and NN listeners perform poorly when listening to NN talkers.