Speech intelligibility in noise with everyday sentences correlates better with matrix sentences than with digits for cochlear implant users and typical hearing listeners
Bram Knipscheer (Leiden University Medical Center)
Hendrik Christiaan Stronks (Universiteit Leiden, Leiden University Medical Center)
Jeroen Johannes Briaire (Leiden University Medical Center)
Johan Hubertus Maria Frijns (TU Delft - Electrical Engineering, Mathematics and Computer Science, Universiteit Leiden, Leiden University Medical Center)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Objective: To determine correlations between speech recognition thresholds (SRTs) for everyday sentences compared to Matrix sentences and digits-in-noise (DIN) triplets. Design: Comparative analysis of SRTs across three speech materials in both cochlear implant (CI) users and typical hearing (TH) listeners, using linear regression and analysis of covariance. Study Sample: 18 experienced CI users (mean age 63 ± 5 years) and 18 age-equivalent TH listeners (mean age 62 ± 12 years), all naive to the test materials. Results: SRTs of Matrix sentences and everyday sentences correlated significantly (R2: 0.81 for CI, 0.71 for TH), as did SRTs of DIN triplets and everyday sentences (R2: 0.42 for CI, 0.28 for TH). Regression slopes did not differ significantly between CI and TH groups in either comparison. However, intercepts differed significantly between the CI (−2.65) and TH (−6.70) groups for the DIN triplets, but not for the Matrix sentences. Slopes for the DIN triplets deviated significantly from unity for both groups. Conclusions: Dutch/Flemish Matrix sentence SRTs closely correlate with everyday sentence SRTs for both CI users and TH listeners, establishing it as a reliable alternative for repeated assessment. DIN triplet SRTs showed weaker correlations with everyday sentences and with significant intercept differences between groups.