Of the 36 million hearing-impaired Americans, 21 million are over 45 years of age, and 35-40% of persons over 65 years of age are hearing-impaired. As the population ages, the number of Americans who are both older and hearing-impaired will only increase. Despite decades of research on speech understanding in noise, there is no theoretical explanation for the increased difficulties older hearing- impaired listener have in multitalker environments as compared to their younger counterparts. This fundamental gap in our knowledge makes it impossible to address clinically the devastating effects of an inability to communicate in social environments such as restaurants, the workplace, and many retail environments. These effects include increased self-perception of handicap and social withdrawal. This research program will seek to determine whether or not speech perception ability in multitalker situations can be predicted by measuring sensitivity to spatial, spectral, ad temporal information. The clinical significance of the research proposed is related directly to the ability to predict how successful an individual listener will be in a given multitalker environment Despite the evidence that binaural hearing, sensitivity to temporal information, and sensitivity to spectral information are impaired in listeners with hearing impairment, and the fact that tests have been developed for the clinic that are intended to be sensitive to these abilities, there are very few clinical audiologists who routinely test these abilities as part of their audiometric examination. If it can be demonstrated which subset of these clinical tests are accurate predictors of speech recognition in a multitalker background, the knowledge gained will allow the clinician to more accurately predict the expected clinical benefit of a particular hearing device for a specific user in various environments. This will allow the clinician to choose and fit hearing aids more successfully and to counsel patients more effectively in terms of the benefit they should expect in various environments. Similar benefits to patient care will be obtained by providing this information to the developers of hearing aid technology and rehabilitative training programs, who could then develop specific devices and techniques to address those deficits that occur most often.
The proposed research will test whether or not individual sensitivity to spatial, spectral, and/or temporal information can predict speech understanding in complex environments. Accurately predicting speech recognition in a multitalker background on the basis of a set of rapid and reliable measures would allow the clinician to design rehabilitatio programs targeting more than simply restoring audibility, which is the only goal currently associated with a simple set of tests: the audiogram. Knowing more precisely the relationship between sensitivity to various cues and predicted benefit would allow the clinician to choose and fit hearing aids more successfully, prescribe rehabilitative auditory training in a more targeted manner, and to counsel patients more effectively in terms of the benefit they should expect.