The effects of high-frequency bandwidth cut-off on speech intelligibility for individuals with high-frequency hearing loss

Date of Completion

January 2010


Health Sciences, Audiology|Health Sciences, Speech Pathology




Theoretically it is advantageous to provide amplification past 4000 Hz since access to high frequencies should improve sound quality and enhance speech perception. However, results have been equivocal within the literature for individuals with sloping high-frequency hearing loss (HFHL) due to insensitive test measures, lack of significant effects, and hearing aid technology having limited bandwidth capabilities. The purpose of this study was to determine if upper cut-off frequency in a receiver-in-the canal (RIC) hearing aid influenced speech perception in quiet, speech perception in noise, and confidence ratings for adults with HFHL. Eighteen adults with HFHL were fit with RIC hearing aids meeting National Acoustics Laboratory Non-Linear (NAL-NL1) targets in three cut-off conditions: 4000, 5500, and 7500 Hz. ^ Speech recognition in quiet was assessed using Pascoe's High-Frequency word list. Speech perception in noise was measured using the Hearing in Noise Test (HINT). Participants rated confidence in responses on Pascoe's list using a four-point Likert scale ranging from Definitely Right to Definitely Wrong. ^ Statistical analyses indicated there was a significant effect of amplification, but the effect of cut-off was minimal. Performance was significantly better for words in quiet for 7500 Hz cut-off versus 4000 Hz. HINT scores revealed no significant difference among aided conditions. Individuals were more confident when using the two highest cut-off conditions in comparison to the 4000 Hz and unaided conditions. Thresholds explained most variance in test measures. Individual participant data suggested that some individuals benefited from the extended cut-off while others did not; a correlation analysis did not reveal significant patterns to explain the variability. ^ Overall results indicated that providing amplification for individuals with HFHL is beneficial in quiet. The results supported a trend for improved performance and confidence with increasing cut-off frequency in quiet, but not in noise. This result suggests that providing additional information above 4000 Hz may not improve speech understanding in background noise; however, no decrement in performance was observed. ^