A-Weighting for All? Acoustic Assessment for an Aurally Diverse Population

Scanlan, Jamie; Davies, William

Abstract

1. The aim of this report is to review whether A-Weighting accurately represents the hearing of the whole UK population, to explore the consequences if it does not, and to make recommendations for future research to address the problems identified.

2. The approach taken was to first research the construction of A-Weighting from its root in early research on loudness perception and the development of the equal loudness contours. We then identified three groups of interest, who may not be well represented by A-Weighting. These are people with age-related hearing loss, neurodivergent people, and noise-sensitive people. These groups were chosen because they each represent a different kind of hearing difference, and they are all a substantial size.

3. A detailed literature review was conducted to examine the available evidence of hearing differences in the three groups and to discuss how these relate to A-Weighting and its uses. We reviewed 540 papers and cite 182 in this report. This dataset was then analysed to identify significant research gaps and suggest future research to fill them.

4. A-Weighting was found to be a significant way in which the assumption of normal hearing is embedded in a very wide range of acoustic assessment and legislation. This is because A-Weighting was developed from the equal loudness contours, and the contours are produced from measurements on a relatively small sample of otologically normal people – 18-25-year-olds with no hearing impairment. This is important because A-Weighting extends far beyond its original use to convert sound levels into loudness estimates and is now found in most assessments and control of noise worldwide, from product design to environmental noise.

5. People with age-related hearing loss represent 7.9% of the UK population. Age-related hearing loss is caused by degradation of the hair cells’ transducer mechanism in the inner ear. Among other effects, this typically causes problems with masking, where it becomes harder to pick a target speech sound out of a noisy background. Because background noise is usually quantified using A-Weighting, and A-Weighting models normal hearing, this may mean that the enhanced need for reduced background interference is often not met for this group.

6. Neurodivergent people (who include autistic people, those with attention deficit hyperactivity disorder (ADHD), dyslexia, dyspraxia, dysgraphia, dyscalculia, and Tourette’s) make up about 10% of the UK population. Evidence was found of auditory processing differences for autistic people, and for people with ADHD and dyslexia. The auditory processing differences occur at a low level in the brain and can result in difficulty separating out sound signals, especially when trying to listen to someone speak against a background of competing speech. It is likely that enhanced background noise targets would improve the experience of this group, but it is currently difficult to translate the laboratory findings on auditory processing into new noise assessment methods. There is too little evidence for people with dyspraxia, dyscalculia, dysgraphia and Tourette’s to base a conclusion on.

7. People who are noise-sensitive represent 17-50% of the population. In this report, this group included people who are more sensitive to general noise, people with misophonia and people with hyperacusis. For this group, the widespread use of A-Weighted sound level to predict noise annoyance is likely to significantly underpredict their heightened response. Hyperacusis is very common in autistic people, and it is particularly likely that A-Weighted levels do not represent the experience of this group.

8. It is not possible to simply adapt the A-Weighting curve to better fit any of the groups at this time. Partly this is because of lack of data on how to adjust, but mostly it is because most of the differences in sound perception are not equivalent to a simple transformation in loudness. The exception is for people with hyperacusis, where more detailed data on their loudness responses might enable a new weighting to be derived.

9. The overarching research gap identified is a lack of translational research. There is a need for research that builds on existing laboratory studies of hearing differences and translates them into guidelines for practitioners to assess sound and design spaces suitable for people with atypical hearing.

10. We also identify nine specific research gaps and projects to address them:
a. Acoustic guidelines for spaces for people with a hearing impairment
b. Acoustic design guidelines for people with auditory processing problems
c. Accurate estimate of the numbers of noise sensitive people in the UK
d. Loudness for people with age-related hearing impairment
e. Loudness functions for autistic people with hyperacusis
f. Equal loudness contours for autistic people with hyperacusis
g. Demographic factors as a modulating effect on neurodivergent hearing
h. Loudness functions for people with hyperacusis
i. Psychoacoustic factors in misophonia
Of these, we suggest that the first three are the most important.

11. This report represents an initial exercise in a much larger project for acoustics to respond to the aural diversity movement – the idea that a wide range of different kinds of hearing differences exist, in contrast to the assumed binary division of normal hearing and impaired hearing. This project could be repeated with other groups of people, such as those with noise-induced hearing loss, Meniere’s disease, or many other conditions.

Citation

Scanlan, J., & Davies, W. (2023). A-Weighting for All? Acoustic Assessment for an Aurally Diverse Population. London: Department for Environment Food & Rural Affairs

• Report Type: Technical Report
• Publication Date: Dec 1, 2023
• Deposit Date: Dec 4, 2024
• Keywords: aural diversity, noise, sound
• Publisher URL: https://randd.defra.gov.uk/ProjectDetails?ProjectId=21660