Starkey Research & Clinical Blog

Can hearing aid settings improve working memory?

Souza, P., & Sirow, L. (2014). Relating working memory to compression parameters in clinically-fit hearing aids. American Journal of Audiology, Just Accepted, released August 14.

This editorial discusses the clinical implications of an independent research study and does not represent the opinions of the original authors.

Working memory provides short-term processing and storage of information during complex cognitive tasks, combining information from numerous sources into a coherent whole (Baddeley, 1992). Incoming stimuli are compared and matched to long-term memory representations, prior to identification and further processing.

The term working memory describes our ability to store and process information during cognitively demanding tasks. In the context of hearing, working memory capacity affects ones ability to match speech inputs with stored representations of that speech. Several studies suggest that individuals with impaired working memory experience increased difficulty understanding speech in complex listening environments (Lunner, 2003). It is assumed that working memory tends to decline with advancing age (Salthouse, 1994). Therefore, it is important to understand how these variables affect speech perception and how they interact with each other, particularly for older hearing aid users.

Working memory may impact the optimal hearing aid characteristics for an individual and a number of studies have investigated the relationship between working memory and wide-dynamic range compression (Foo, et al., 2007; Gatehouse, et al., 2006; Lunner & Sundewall-Thoren, 2007; Ohlenforst, et al., 2014).  In these studies,  hearing aid compression speed was examined while keeping other amplification characteristics constant. Subjects with better working memory were generally found to perform better with fast-acting compression, whereas subjects with poorer working memory performed better with slow-acting compression.  The authors interpret these results as an indication that fast-acting compression alters the speech envelope in ways that make it more difficult to match incoming stimuli to stored lexical representations (Jenstad & Souza, 2007; Jenstad & Souza, 2005; Ronnberg et al., 2013; Ronnberg et al., 2008).

Laboratory experiments inherently must control the variables under study in order to glean meaningful interpretations. However, comparing fast and slow compression speed in isolation does not represent the typical conditions of a clinical hearing aid fitting, in which these characteristics are not independently adjustable. Furthermore, with changes in compression speed from one hearing aid model to another, many other variables are likely to differ as well, such as feedback management, noise reduction characteristics and the number of compression channels. These considerations make it difficult to extrapolate laboratory findings to everyday clinical experiences. The goal of Souza and Sirow’s study was to examine how compression speed and working memory relate to each other, using selection, fitting and verification techniques as they would typically be used in a clinical setting.

Twenty-seven participants with hearing loss were fitted with at least three different sets of receiver-in-canal hearing instruments, from several manufacturers. Because only one manufacturer offered an aid with adjustable compression speed, each subject completed a comparison of two compression settings with this single hearing aid, plus two or three additional models from other manufacturers that varied in their compression characteristics.  All aids were fitted with closed domes in the appropriate size for the individual. Real-ear verification and adjustments to prescribed levels were completed as they would in a typical clinical hearing aid fitting, to ensure audibility and comfort.  Aids were programmed with omnidirectional microphones and special hearing aid parameters such as feedback management and noise reduction were set according to manufacturer defaults.

Working memory is often assessed with a dual-paradigm task, in which the subject is required to process information while storing it for later recall. In this study, working memory was assessed with a reading span test, the same procedure used in previous studies of hearing aid compression and working memory. Subjects were presented with five-word sentences flashed on a computer screen and were asked to judge if the sentences made sense or not.  Sentences were presented one at a time in blocks of three, four or five. After each block, subjects were asked to recall either the first or last words of the sentences. The working memory score was taken as the percentage of correctly recalled words across all blocks.

Speech intelligibility was tested using the QuickSIN (Killion, et al., 2004) test, because of its ease of clinical administration and similarity to test materials and conditions in prior studies of compression (Lunner & Sundelwall-Thoren, 2007).  The test was administered in a sound booth via loudspeaker at a 0-degree azimuth, at 70dBHL for most subjects.  The QuickSIN yields an SNR loss score, which indicates the increase in SNR required to achieve a performance threshold. Larger SNR loss scores represent poorer performance.

Correlations were calculated to examine the relations among working memory, age, degree of hearing loss and speech-in-noise performance.  Not surprisingly, increases in age and degree of hearing loss were associated with poorer scores on the QuickSIN test. Working memory scores were also significantly correlated with aided QuickSIN scores.  Lower working memory scores were loosely associated with increased age and poorer unaided QuickSIN scores, but these relationships did not reach significance.

Reading span test scores, representing working memory, ranged from 17% to 50%, with a mean of 34%.  As in previous studies, subjects were divided into high and low working memory groups, based on the median score for the group. For slower compression speeds, comparable performance was achieved by both high and low working memory groups. At faster compression speeds, individuals in the high working memory group performed better than those in the low working memory group. For the fastest compression times, the difference in SNR loss between the high and low working memory groups was greater than 5dB. The authors point out that this is a substantial difference, as a QuickSIN SNR loss difference of 2.7dB is considered significant.

Aided QuickSIN scores were significantly affected by working memory for fast compression speed, but not for slow compression speed. There was high variability in the scores, especially for slow compression times, so further analysis was conducted to examine the contributions of other variables. For fast compression speeds, working memory and hearing loss accounted for most of the variance. For slow compression speeds, age and hearing loss were significant predictors of performance, but working memory was not.

The results of this study are consistent with previous reports suggesting that listeners with low working memory may not perform well with fast acting compression, whereas those with high working memory can be expected to do better.  The findings of the current study appear particularly robust because they emerged under less controlled conditions than in the laboratory studies. The authors point out that even in the hearing aid that allowed manipulation of compression speed, changing it resulted in other changes in signal processing as well.  The fact that compression speed still had a significant effect on speech-in-noise performance under these conditions is support for its relationship with working memory.

Though further study is needed to illuminate the relationship between working memory and the selection of hearing aid parameters, there are a number of potential benefits to incorporating working memory tests into clinical practice. The working memory assessment could help to explain poor performance with a current set of hearing aids and indicate the need for new aids or adjustments to signal processing parameters, if possible. Souza and Sirow offer a cautionary statement regarding the use of working memory assessment during a diagnostic hearing evaluation. They suggest that patients may not understand the link between auditory assessment and a task that could involve assessment of memory. With that cautionary consideration, hearing care providers may be more likely than other clinical professionals to recognize symptoms of cognitive decline. Atypical results of a working memory assessment may provide insight into a patient’s performance with hearing aids as well as their general cognitive health, prompting referrals to a primary care physician or other specialists.

The study of hearing loss, hearing aids, cognition and memory is an interesting area of inquiry with potentially important implications for clinical hearing aid fitting. Souza and Sirow’s report on the relationship between working memory and compression speed illustrates how individual variability in working memory could have specific impact on the selection of hearing aid characteristics.  Their findings represent an important link between laboratory investigation on this topic and the clinical prescription of hearing aids.


Baddeley, A. (1992). Working memory. Science 255 (5044), 556-559.

Foo, C., Rudner, M., Ronnberg, J. & Lunner, T. (2007). Recognition of speech in noise with new hearing instrument compression release settings requires explicit cognitive storage and processing capacity. Journal of the American Academy of Audiology 18(7), 618-631.

Gatehouse, S., Naylor, G. & Elberling, C. (2006). Linear and nonlinear hearing aid fittings: 2. Patterns of candidature. International Journal of Audiology 45(3), 153-171.

Jenstad, L. & Souza, P. (2005). Quantifying the effect of compression hearing aid release time on speech acoustics and intelligibility. Journal of Speech, Language and Hearing Research 48(3), 651-667.

Jenstad, L. & Souza, P. (2007). Temporal envelope changes of compression and speech rate: combined effects on recognition for older adults. Journal of Speech, Language and Hearing Research 50(5), 1123-1138.

Killion, M., Niquette, P., Gudmundsen, G., Revit, L. & Banerjee, S. (2004). Development of a quick speech-in-noise test for measuring signal-to-noise ratio loss in normal-hearing and hearing-impaired listeners. Journal of the Acoustical society of America 116(4), 2395-2405.

Lunner, T. (2003). Cognitive function in relation to hearing aid use. International Journal of Audiology 42 (Suppl.): S49-S58.

Lunner, T. & Sundewall-Thoren, E. (2007). Interactions between cognition, compression and listening conditions: effects on speech-in-noise performance in a two-channel hearing aid. Journal of the American Academy of Audiology 18(7), 604-617.

Ohlenforst, B., Souza, P. & MacDonald, E. (2014). Interaction of working memory, compressor speed and background noise characteristics. Paper presented at the American Auditory Society, Scottsdale, AZ.

Remensnyder, L. (2012). Audiologists as gatekeepers and it’s not just for hearing loss. Audiology Today, July/August,  24-31.

Ronnberg, J., Rudner, M., Foo, C. & Lunner, T. (2008 ). Cognition counts: a working memory system for ease of language understanding (ELU). International Journal of Audiology 47(Suppl. 2), S99-105.

Ronnberg, J., Lunner, T., Zekveld, A., Sorqvist, P., Danielsson, H., Lyxell, B. & Rudner, M. (2013). The Ease of Language Understanding (ELU) model: theoretical, empirical and clinical advances. Frontiers in Systems Neuroscience 7, 31.

Rudner, M., Foo, C., Ronnberg, J. & Lunner, T. (2009). Cognition and aided speech recognition in noise: specific role for cognitive factors following nine-week experience with adjusted compression settings in hearing aids. Scandinavian Journal of Psychology 50(5), 405-418.

Salthouse, T. (1994). The aging of working memory. Neuropsychology 8(4), 535-543.