You are in: Home » History of the Unit
2. HEARING
General Notes. This material has been scanned from the original typescript, while we have done our best to remove errors, some may well remain. You can access other parts of this particular Progress Report either from the menu at the bottom of this entry or by navigating back to the Unit history timeline. Reference for this report are indexed by number and these can be found in a dedicated section also accessible from the menu at the bottom of this entry.
2.1 Noise disruption of auditory processing (Barton, Johnson-Davies Lutfi, Milroy, Nimmo-Smith, R. Patterson, Weber)
This project aims to produce a mathematical model of auditory aasking capable of predicting when an interfering background noise will make an important signal difficult or impossible to hear. The primary problem is to specify the frequency-resolving power of the ear with a set of accurate, but convenient, equations. We are pursuing two approaches to this problem. The first involves describing the frequency resolution in terms of an auditory filter shape; it is a direct extension of our earlier work. The second involves specifying the frequency resolution in terms of an auditory excitation pattern and it represents a new and complementary approach to the problem.
When a listener is trying to hear a signal in the presence of a noise background, he behaves as if he were centering a filter on the signal to improve its detectability. The auditory filter passes the signal and progressively rejects the noise as the distance of the noise component from the centre frequency of the filter increases. The ability to predict masking depends entirely on the accuracy with which we can specify the attenuation characteristic or shape of the auditory filter (178). The filter is not always symmetric and at times its centre is shifted away from the signal a little to improve detection. Until recently the interaction of these two phenomena limited the accuracy of our filter-shape estimates. To alleviate this problem, Nimmo-Smith has developed a completely new mathematical analysis which has disposed of most of the restrictive assumptions required by earlier methods. The success of this new approach was demonstrated by Patterson and Nimmo-Smith (184) and it is now being used to extend the generality of the model to high intensity noise backgrounds (124 U; 125 U) and to a wider population of listeners (185 U). The method has also been adopted in several other laboratories to extend the frequency range of the model.
The alternative approach to the specification of auditory frequency selectivity has the advantage that it is more directly related to the physiological mechanisms underlying the process. It has the disadvantage, however, that the appropriate experiments required to create a competent, quantitative model of masking are much more difficult to perform. This approach involves the use of electrodes to measure neural processes in animals, followed by an attempt to generate similar data behaviourally with humans. Our initial investigations showed that the analogy between animal and human experiments has serious flaws (114 U; 115).
Since then we have concentrated our efforts on determining how the psychophysical experiments could be changed to improve the analogy with the animal data (237; 239 U).
2.2 The design and evaluation of auditory warning systems (Barton,Milroy, Nimmo-Smith, Patterson)
With increasing technology, there is a growing use of machines to monitor for dangerous conditions. When a problem is detected, the incident is typically signalled using an auditory warning, since its effectiveness does not depend on where the observer is looking. We are using the auditory filter-model of masking to develop a set of principles for designing auditory warnings for specific noise backgrounds. Currently most of our work on this project involves the auditory warnings used on the flight decks of civil aircraft. We have so far established the appropriate sound levels, and investigated the relationship between number of warnings and probability of confusion between them (180; 182). On the basis of this research we were asked to review the auditory warning set proposed for a new aircraft (183) before going on to design a prototype advanced warning system (177).
2.3 Improved methods of audiometric assessment (Milroy, Nimmo-Smith, Patterson, Weber)
The traditional hearing test, the audiogram, provides only a rudimentary evaluation of hearing, a specification of how intense a sound must be to get into the system without regard for the fidelity of the system. It seems probable that the auditory filter measure of frequency resolution might provide a better estimate of a listener's ability to process sound effectively. Hearing deteriorates with age, and so before proceeding to work involving hearing-impaired patients, we have performed a survey study to determine how the auditory filter shape changes with age (181). This initial study showed that there is indeed a significant broadening of the filter for people beyond age 60. In addition, it showed that the shape of the filter changes and this has led us to choose a new class of functions to represent the auditory filter (185 U). Currently this work has been extended, using the method of Patterson and Nimmo-Smith to assess the asymmetry of these broader filters and to determine whether listeners with broad filters have reduced ability to understand speech.