Speech Perception in Noise and Listening Effort of Older Adults With Nonlinear Frequency Compression Hearing Aids

Abstract

Objectives: The purpose of this laboratory-based study was to compare the efficacy of two hearing aid fittings with and without nonlinear frequency compression, implemented within commercially available hearing aids. Previous research regarding the utility of nonlinear frequency compression has revealed conflicting results for speech recognition, marked by high individual variability. Individual differences in auditory function and cognitive abilities, specifically hearing loss slope and working memory, may contribute to aided performance. The first aim of the study was to determine the effect of nonlinear frequency compression on aided speech recognition in noise and listening effort using a dual-task test paradigm. The hypothesis, based on the Ease of Language Understanding model, was that nonlinear frequency compression would improve speech recognition in noise and decrease listening effort. The second aim of the study was to determine if listener variables of hearing loss slope, working memory capacity, and age would predict performance with nonlinear frequency compression.

Design: A total of 17 adults (age, 57-85 years) with symmetrical sensorineural hearing loss were tested in the sound field using hearing aids fit to target (NAL-NL2). Participants were recruited with a range of hearing loss severities and slopes. A within-subjects, single-blinded design was used to compare performance with and without nonlinear frequency compression. Speech recognition in noise and listening effort were measured by adapting the Revised Speech in Noise Test into a dual-task paradigm. Participants were required trial-by-trial to repeat the last word of each sentence presented in speech babble and then recall the sentence-ending words after every block of six sentences. Half of the sentences were rich in context for the recognition of the final word of each sentence, and half were neutral in context. Extrinsic factors of sentence context and nonlinear frequency compression were manipulated, and intrinsic factors of hearing loss slope, working memory capacity, and age were measured to determine which participant factors were associated with benefit from nonlinear frequency compression.

Results: On average, speech recognition in noise performance significantly improved with the use of nonlinear frequency compression. Individuals with steeply sloping hearing loss received more recognition benefit. Recall performance also significantly improved at the group level, with nonlinear frequency compression revealing reduced listening effort. The older participants within the study cohort received less recall benefit than the younger participants. The benefits of nonlinear frequency compression for speech recognition and listening effort did not correlate with each other, suggesting separable sources of benefit for these outcome measures.

Conclusions: Improvements of speech recognition in noise and reduced listening effort indicate that adult hearing aid users can receive benefit from nonlinear frequency compression in a noisy environment, with the amount of benefit varying across individuals and across outcome measures. Evidence supports individualized selection of nonlinear frequency compression, with results suggesting benefits in speech recognition for individuals with steeply sloping hearing losses and in listening effort for younger individuals. Future research is indicated with a larger data set on the dual-task paradigm as a potential cognitive outcome measure.

Figure 1

The average and range of behavioral pure tone air conduction thresholds.

Figure 2

Boxplots for percent correct sentence-ending word recognition in noise (A) and word recall (B) with non-linear frequency compression off and on. Scores were averaged across high and low contexts.

Figure 3

Percent of sentence-ending words correctly recognized (A) and recalled (B) with and without non-linear frequency compression (NFC) enabled. Participant scores are ordered from least recognition benefit to most recognition benefit with non-linear frequency compression.

Figure 4

Correlation between the change in recognition performance with and without non-linear frequency compression and slope of hearing loss, calculated as the threshold slope between 1–4 kHz.

Figure 5

Correlation between the change in recall performance with and without non-linear frequency compression and age in years.

Figure 6

Average pure tone thresholds for individuals who received the least benefit (-9% to +5% change; Group A) and the most benefit (+5% to +22% change; Group B) for speech recognition performance with non-linear frequency compression in low context sentences.