Effect of spatial separation and noise type on sentence recognition by Mandarin-speaking cochlear implant users

Abstract

Objectives: To investigate the effects of spatial separation and noise type on sentence recognition by unilateral Mandarin-speaking cochlear implant (CI) users and normal-hearing (NH) listeners.Method: Twenty-two unilateral Mandarin-speaking CI users and six NH listeners participated in this study. Speech reception thresholds were measured for three noise types (steady state noise, speech babble, and music). Sentences from the Mandarin Speech Perception test were presented directly in front of the listener (0°). Noise was presented from one of the five speaker locations: −90°, −45°, 0°, +45°, and +90°.Results: Overall, CI performance was significantly poorer than NH performance for all spatial separation and noise type conditions. NH listeners performed best with music and poorest with steady noise. CI users performed best with steady noise, and poorest with babble. Performance was significantly affected by noise location and noise type. There was no significant difference in head shadow effects among the different noise types for CI users.Conclusions: Performance was much poorer in CI than in NH listeners for all noise types and spatial separations. Noise type differently affected unilateral CI users and NH listeners. The limited spectral resolution in CI users did not appear to affect head shadow.