The impact of abnormal frequency selectivity on speech recognition is the focus of the proposed research. Frequency selectivity refers to the ability of the auditory system to separate the various frequency components of complex sounds, and is estimated by measuring the change in the threshold of audibility of a sound that results from the presence of another sound. Using techniques that facilitate comparisons of performance among listeners whose absolute thresholds vary in magnitude and configuration, the goals of the proposed experiments are (1) to demonstrate that, under certain conditions, abnormal frequency selectivity results in a significant disruption in speech recognition for hearing-impaired listeners, and (2) to examine the mechanisms responsible for this disruption. To account for the absence of a strong association between frequency selectivity and speech recognition in previous investigations, Experiment I determines if speech recognition for hearing-impaired listeners is poorer than normal when measured in the presence of maskers that yield abnormally-high masked thresholds for these listeners. Experiment II determines if elevated masked thresholds resulting from increased masker levels interfere to a greater extent with speech recognition for hearing-impaired individuals than normal-hearing listeners. The mechanism of suppression, in which strong activity at one frequency suppresses weaker activity at adjacent frequencies, is examined in Experiments III and IV. Experiment III is designed to determine if suppression enhances speech recognition in noise for normal-hearing listeners by providing a more advantageous signal-to-noise ratio. Experiment IV determines if suppression is reduced or absent in listeners with cochlear hearing loss, and if the absence of suppression interferes with speech recognition in noise for these listeners. From the results of these experiments, a better understanding of the impact of abnormal frequency selectivity on speech recognition will emerge, and will facilitate development of more appropriate signal-processing schemes for devices that are designed to address the receptive communication difficulties imposed by sensorineural hearing loss.
Showing the most recent 10 out of 69 publications
