The long-term objective of this project is to provide a better understanding of the processes underlying sound source segregation in the normal auditory system and the effects of cochlear dysfunction on those processes. This will contribute, in particular, to an objective appreciation of the difficulties experienced by listeners with cochlear hearing loss in competitive listening situations. Four areas of study are proposed. The first area deals with the manner in which information concerning a spectrally-dispersed signal is combined across frequency to improve detection. Of interest is a comparison between conditions where the signal consists of an increment in energy across frequency and conditions where the signal consists of a decrement in energy across frequency. The second area of study is concerned with situations where a complex signal must be extracted from a background containing multiple modulation patterns.
The aim i s to determine the degree to which the auditory system can implement concurrent across-frequency processes to extract the complex signal and to ascertain the limitations imposed by impairments of frequency and temporal resolution on this ability. The third area of study examines the correspondence between measures of across-frequency processing and auditory grouping. One series of experiments assesses the susceptibility of comodulation masking release (CMR) and modulation detection interference (MDI) to factors which oppose the formation of auditory groupings. A second investigation implements a CMR paradigm to provide an objective measure of the fusion strength of purported auditory grouping factors. The fourth area of study quantifies the ability of cochlear-impaired listeners to segregate simultaneously-present sounds, and to segregate (or stream) sequentially-present sounds. Of primary interest is the relationship between these abilities and the functions of frequency selectivity and temporal resolution in the same listeners. A total of 15 cochlear-impaired listeners and a control group of 12 normally-hearing listeners will be recruited using strict age and audiometric criteria. Testing will take place in a sound-treated room under headphones using computer-controlled adaptive procedures. Data will be analyzed using simple descriptive statistics, correlations, and analyses of variance.
Showing the most recent 10 out of 34 publications
