Physiologically inspired simulation of sensorineural hearing loss

With an ageing population, hearing impairment is becoming an increasingly common problem. The most common complaint of hearing impaired listeners is that it is difficult to understand speech in background noise, but, surprisingly, the degree to which listeners struggle with speech in noise is not predictable from their degree of hearing loss.

Hearing aids can help people to hear sound better in quiet environments, but, in the presence of background noise, they are of little use. Advances in hearing aids are limited by our lack of basic understanding of hearing impairment. The ear is a very complex organ, which we know can go wrong in a variety of ways. Unfortunately, no methods currently exist to determine what exactly is wrong with the inner ear of a given hearing-impaired person. People with hearing impairment seem to have a variety of symptoms, that are probably caused by different combinations of underlying inner-ear problems.

This project will construct sounds that simulate the auditory experience associated with different types of hearing impairment. These sounds will be presented to normally-hearing listeners to test performance in different auditory tasks. It will then be possible to determine which underlying physical problems give rise to the set of symptoms that each patient experiences, giving insights into their underlying condition. The main purpose of the present research, therefore, is to develop the simulation method and to demonstrate that it can reproduce the patterns of problems experienced by hearing impaired listeners.

Simulations of hearing loss would be useful in a number of ways.

First, the simulations will enable researchers to link underlying causes of hearing impairment with characteristic patterns of hearing difficulty. This will contribute to an ongoing major shift towards diagnosing hearing loss in terms of undelying pathologies rather than symptoms.

Second, using a simulation, it will be possible to test new hearing aid designs using listeners with normal hearing. The main benefit of this is that, for each test, listeners would all have exactly the same simulated hearing impairment rather than the unknown mixture of conditions that hearing impaired people might have. The effectiveness of the hearing aid in a particular task and for a particular condition can thus be established.

Third, simulations will be valuable to demonstrate to normally hearing listeners (relatives, audiology students etc.) the nature and experience of hearing impairment. Moore and colleagues (1995) have produced a CD, and Zurek and Desloge (2007) have produced a commercial simulator for the purpose of audiological demonstrations. These simulations were based on the phenomenology rather than the physiology of hearing impairment. The benefit of physiologically inspired simulations is that normally hearing listeners can appreciate the problems experienced by hearing-impaired individuals with particular hearing pathologies.

The simulation software and vocoder software will be posted on our website in a ready-to-use form.

The simulation technique will, for the first time, allow researchers to simulate perceptual effects of specific forms of sensorineural hearing impairment. The simulation of specific physiological impairments will reveal the characteristic patterns of perceptual deficit that different pathologies produce, thus opening a path to differential diagnosis of sensorineural hearing loss.

The technique will facilitate, for the first time, the independent comparative assessments of commercial hearing aids and hearing-aid amplification strategies for specific hearing pathologies.

The combination of the vocoder with hearing aid simulations will make it possible for hearing aid manufacturers to develop amplification strategies that specifically address different aspects of hearing impairment. This will facilitate the future development of bespoke amplification strategies for patients with different types and combinations of impairment.

Building on Moore's (1995) work and Zurek's (2007) work, we will produce a series of simulations based on the new vocoder that will illustrate the effects of different forms of hearing impairment. The new simulations will reflect the increased understanding that has developed over the last 20 years or more, producing a wider range of simulated effects that are rooted more in the underlying physiology. Rather than producing and marketing a CD, we will place these demonstrations on a web page. This would be of value in the training of audiologists in order to deepen their insight into hearing difficulties. It will also be of use as a demonstration tool for family members of hearing-impaired individuals and for charities as an illustration of the difficulties faced by hearing-impaired people.