Our long-term objective is to reach an understanding of the auditory system that will allow quantitative modeling of the perception of tones, noise, and speech in both normal and impaired hearing. Such models are important for understanding normal auditory processes and provide a basis for improving diagnosis and rehabilitation of hearing-impaired listeners. The proposed project aims to obtain psychoacoustic data that will allow further tests of the concept that excitation patterns and multiband decision rules may provide a general framework for quantitative modeling of normal and impaired hearing. Based on 21,2AFC experiments in normal, impaired, and masked-normal listeners, the proposed project will establish a firm basis for refining our excitation-pattern model of intensity discrimination and testing a number of hypotheses related to the decision process In this and other models.
The specific aims are: (1) To test the hypothesis that differences in across-frequency integration for detection of brief and long tone complexes reflect different psychometric functions, psychometric functions will be measured for detection of masked brief and long tones and tone complexes. (2) To test the hypothesis that lower-than-expected thresholds for brief tone complexes depends on synchronous activity across a number of channels, masked thresholds will be measured for brief tone bursts and tone complex with the component tones in various phase relationships at the envelope peak. (3) To test the hypothesis that phase-locking of the auditory nerve may aid detection of tones with long durations, thresholds in the quiet and under masking will be compared for long and short bursts of tones and frozen narrow-band noises at several frequencies. (4) To test the hypothesis that the effect of duration on level discrimination depends on the width of the excitation pattern, level discrimination thresholds will be measured for tones and tone complexes partially masked by notched-noise maskers. (5) To test the hypothesis, that detection of masked signals is based on activity in an optimum selection of equally weighted frequency-selective channels, detection-probability functions (Conditional On a Single Stimulus, COSS) will be measured for pure tones and an 18-tone complex in either fixed or random conditions. The signals will be masked by uniformly masking noise composed of 24 critical-band wide noises whose levels vary randomly. The conditional psychometric functions thus obtained provide direct estimates of the decision rules used for each signal and listening condition. (6) To test the hypothesis that listeners' decision in a level-discrimination experiment is based on an optimally weighted sum of across-interval excitation-level differences in frequency selective channels, COSS functions will be measured for level discrimination of multitone complexes in which a random jitter is applied to each component. These data will add significantly to our understanding of normal and impaired hearing and will be important for the development and testing of models of the auditory system. Little is known about how the auditory system integrates information across channels, although this ability is critical to our perception of almost all sounds.
Showing the most recent 10 out of 24 publications
