Speech perception in modulated noise assessed in bimodal CI users.

Abstract

Although good speech perception in quiet is achievable with cochlear implants (CIs), speech perception in noise is severely impaired compared to normal hearing (NH). In the case of abimodal CI fitting with ahearing aid (HA) in the opposite ear, the amount of residual acoustic hearing influences speech perception in noise. The aim of this work was to investigate speech perception in noise in agroup of bimodal CI users and compare the results to age-matched HA users and people without subjective hearing loss, as well as with ayoung NH group. Study participants comprised 19bimodal CI users, 39HA users, and 40subjectively NH subjects in the age group 60-90years and 14young NH subjects. Speech reception thresholds (SRTs) in noise were adaptively measured using the Oldenburg Sentence Test for the two spatial test conditions S0N0 (speech and noise from the front) and multisource-noise field (MSNF; speech from the front, four spatially distributed noise sources) in continuous noise of the Oldenburg Sentence Test (Ol-noise) and amplitude-modulated Fastl noise (Fastl-noise). With increasing hearing loss, the median SRT worsened significantly in all conditions. In test condition S0N0, the SRT of the CI group was 5.6 dB worse in Ol-noise than in the young NH group (mean age 26.4years) and 22.5 dB worse in Fastl-noise; in MSNF, the differences were 6.6 dB (Ol-noise) and 17.3 dB (Fastl-noise), respectively. In the young NH group, median SRT in condition S0N0 improved by 11 dB due to gap listening; in the older NH group, SRTs improved by only 3.1 dB. In the HA and bimodal CI groups there was no gap listening effect and SRTs in Fastl-noise were even worse than in Ol-noise. With increasing hearing loss, speech perception in modulated noise is even more impaired than in continuous noise.