Processing of prosodically marked focus in a cochlear implant simulation by non-native listeners: Preliminary results

Abstract

Speakers can use prosodic cues to direct listeners to a specific part of an utterance. The prosodically emphasised part has linguistic focus, determined by the semantic and pragmatic context (e.g., Cole, 2015). For cochlear implant (CI) users, processing prosodically marked focus can be challenging given the degradation in fine spectrotemporal detail of the signal transmitted through the device (e.g., Başkent et al., 2016). An additional challenge can be expected for CI users listening to a non-native language. In this ongoing study, we investigate how native Dutch learners of English process prosodically marked focus in English sentences degraded by a CI simulation compared to how they process it in non-CI-simulated stimuli. These results are compared to those of native English listeners. Listeners are presented with English sentences differing in prosodically marked sentential focus and are instructed to indicate which of four possible context questions prompted the response stimulus. We expect that listeners are less accurate and less efficient for the CI-simulated stimuli compared to the non-CI-simulated stimuli and that non-native listeners are less efficient than native listeners, underlining the challenges of prosodically marked sentential focus processing in a non-native language with CI hearing.