11th Speech in Noise Workshop, 10-11 January 2019, Ghent, BE

Listening effort during non-native speech perception in noise

Giulia Borghini(a), Valerie Hazan(b)
Department of Speech Hearing and Phonetic Sciences, University College London, United Kingdom

(a) Presenting
(b) Attending

Current evidence demonstrates that the presence of a background noise is much more detrimental for non-native than for native listeners in terms of performance. Moreover, we know that listening effort is increased for non-native compared to native listeners, even when the intelligibility levels are equated across the two groups.
Here, I discuss results from a pupillometry study exploring listening effort during a speech perception task in noise. The listening task featured lists of 12 semantically related and unrelated sentences. In order to investigate the contribution of semantic context to listening effort, results were analysed by grouping sentences into three parts (beginning, middle and end section). Participants included 21 Italian learners of English, and 18 native English controls. An adaptive procedure was used to match the intelligibility levels across participants. An intelligibility level of 80% was targeted.

Preliminary analyses consistently confirmed that non-native listeners showed a significantly greater pupil response compared to native listeners when attending to speech in noise. Additionally, we only found a significant reduction in the pupil peak dilation when native participants were attending to related compared to unrelated sentences. This benefit from context was not replicated for the non-native listeners. Results suggest that non-native listeners do not benefit in terms of reduction in the listening effort as native listener do when they are provided with a consistent semantic context.

Last modified 2019-01-08 16:51:41