The Role of the Efferent System in Auditory Gain Reduction

The ability of the human ear to detect small differences in sound level relies on the response properties of the basilar membrane (BM) within the cochlea of the inner ear. When sound waves enter the cochlea, they cause the BM to vibrate, with each place on the membrane responding most to a sound of a particular frequency (its characteristic frequency ). The BM response is amplified for sounds with a frequency close to that of the characteristic frequency. This amplification or gain varies with the level of a sound, with more gain applied to the BM response to low-level sounds than to higher-level sounds. Since the amount of gain applied in response to a sound varies with its level, the applied gain is described as level-dependent . Although we now understand to some extent how gain varies with sound level, we still do not fully understand how gain varies with the duration of a sound. Behavioural studies on humans suggest that the gain is greatest at the start of a sound compared to the end of a sound. This reduction of gain is thought to be due to neural feedback of information from higher auditory centres in the brain, and is often referred to as the efferent effect. Previous behavioural studies on humans involving sound presentation by headphones, have attempted to measure the gain variation with sound duration using a technique called simultaneous masking . In this method the detectability of a sound (signal) is measured in the presence of a masking sound (masker) presented at the same time. If the signal and masker are of the same frequency and signal detectability is measured for a range of sound levels, an estimate of the change in gain at the characteristic frequency of the BM can be obtained for different masker durations. However the interpretation of such studies may be confounded by the presence of other complex interactions on the BM and so may not reflect an accurate estimate of the gain and efferent effect. The present study uses a technique called the additivity of masking which involves forward masking . In forward masking signal detectability is measured for a signal presented soon after a masker. Since the signal and masker are not present at the same time, this method reduces the involvement of other complex interactions along the BM which could affect the true estimate of the gain. The present proposal will use the additivity of masking technique to find i) how the gain increases and decreases with the duration of a sound; ii) If the frequency of the sound affects how the gain varies with duration; ii) How increases in sound level affect the variation of gain with sound duration. The findings will increase our understanding of the efferent effect in normal-hearing individuals. One key area that would benefit from an increased knowledge of the human efferent effect is that of the treatment of hearing impairment. The efferent effect is vital for detecting and understanding speech. However many individuals with hearing impairment lack an efferent effect. By obtaining accurate estimates of the efferent effect for a range of auditory stimuli we will be able to improve signal processing strategies for hearing aids and cochlear implants.

The Proposed Research will contribute to the Following Scientific Advances in Knowledge: (i) Measurement of the temporal build-up and decay of the human efferent effect. (ii) Estimate of the frequency range of the efferent effect. (iii) Measurement of the duration of the efferent effect. (iv) Investigation of the combined effects of precursor frequency, level, duration and onset, on the efferent effect. The Proposed Research will Contribute to Advances in Methodology & Techniques by use of: (i) An integrative approach including psychophysics and modelling to understand the human efferent response. (ii) A novel variation of the AFM technique to obtain an estimate of peripheral gain reduction under efferent stimulation. The AFM technique avoids i) the contribution of suppression to the gain estimates and ii) the need for a linear response reference (off-frequency masker). The key beneficiaries to gain from the proposed research are: (i) Cochlear Implantees: One key area that would benefit from a detailed characterisation of the human efferent (or temporal) effect is that of bilateral cochlear implantation and evaluation; preliminary attempts to restore the temporal effect in cochlear implantees have been shown to improve speech performance. (ii) Individuals with an Auditory Processing Disorder (APD): Recent studies suggest that reduced MOCB activity may contribute to some of the hearing difficulties encountered by children with an APD, with a number of physiological studies suggesting that the efferent system plays a key role in hearing in background noise. The present set of experiments will contribute to an improved understanding (level-, duration- and frequency- dependency) of the efferent response in normal-hearing individuals and indicate ways in which tests could be designed as part of a clinical test for early identification of individuals with APD. (iii) Clinical Sector: The advantage of an accurate psychophysical test of efferent function (as compared to otoacoustic emissions) is that it provides a measure of the degree of efferent function, and for that reason allows for the first time a categorisation of different levels of efferent function. (iv) Technological sector: The research findings will have implications for improved modelling of the auditory periphery, such as modifying the hearing aid compression over time to simulate the efferent response. Such modified models could serve as the front-end to speech-recognition systems. (v) Academic/research sector: The utility and accuracy of models of the human auditory system will be greatly enhanced by simulatation of the action of the efferent effect. (vi) Health and Safety Executive (HSE): The efferent system may have a role to play in protecting the auditory system from noise-induced hearing loss. There would be great interest in having a psychophysical test that could be use clinically to obtain a measure of the robustness of efferent function in order to predict susceptibility to a NIHL. The overall benefits to society of a comprehensive audiological assessment programme including efferent assessment are: i) Improved auditory assessment and diagnosis ii) An overall reduction in cost of clinical staff time treating individuals iii) Early detection of the effects of occupational noise exposure iv) Reduced compensation claims v) Improved productivity in the workplace