Effects of Slow- and Fast-Acting Compression on Hearing-Impaired Listeners' Consonant-Vowel Identification in Interrupted Noise.

Only nine hearing-impaired listeners completed the aided conditions, and they were older adults with mild to moderately severe high-frequency loss, so results may not apply to other hearing losses or ages. The normal-hearing and hearing-impaired groups differed by nearly 50 years in age, so age and hearing loss cannot be separated. Speech material was nonsense consonant-vowel tokens, not words or sentences, and the noise was artificial interrupted white noise, so real-world listening is not tested. Hearing aids were simulated with fixed compression ratios and a shadow-controlled quiet condition, not actual hearing aids in everyday environments. Linear amplification was included as a baseline but provided unrealistically high gain at high input levels. There was large variability among hearing-impaired listeners, and some did not benefit from higher audibility. Ceiling effects may have hidden differences at higher speech levels. No cognitive assessment was done, and context or top-down processing was prevented by nonsense tokens.

Funders named were Danmarks Tekniske Universitet and Sonova AG. The supplied text does not say whether either funder designed the study, collected data, or wrote the manuscript.

Do not read the 8% to 9% improvement as proof that fast compression will help patients understand everyday speech. It came from nine hearing-impaired listeners identifying nonsense consonant-vowel tokens through a simulator in artificial interrupted noise. The benefit was not present at the lowest or highest speech levels.