Frequency modulation detection, frequency selectivity, and speech recognition in noise by hearing‐impaired listeners.

Abstract

Reduced ability to make use of temporal fine structure (TFS) information may contribute to deficits among hearing impaired (HI) listeners for speech recognition in competing background sounds. The broad tuning typically observed in HI listeners will produce more complex patterns of TFS as input to the auditory nerve, and these patterns may be more difficult to encode and/or interpret by higher auditory centers. In the current study, we tested HI listeners on a frequency‐modulation (FM) detection task thought to require the use of TFS cues, on tone detection in notched‐noise to assess frequency selectivity, and on speech recognition in continuous and modulated background sounds. Performance on the psychoacoustic tasks was tested at four frequencies (500, 1000, 2000, and 4000 Hz) at levels between 60 and 85 dB SPL (approximating the range of third‐octave band levels in the speech stimuli). HI listeners with similar audiometric thresholds showed clear differences in frequency tuning, FM detection thresholds, and speech performance. Relationships between the psychoacoustic measures in the four frequency regions and between these measures and speech performance will be discussed.