ANALYSIS OF THE PERFORMANCE OF A MODEL-BASED OPTIMAL AUDITORY SIGNAL PROCESSOR

Traditionally, psychophysical data have been predicted either by const ructing models of the peripheral auditory system or by applying signal detection theory (SDT). Frequently, the theoretical detection perform ance predicted by SDT is greater than that observed experimentally and a nonphysiologically based ''internal noise'' source is often added t o the system to compensate for the discrepancy. A more appropriate exp lanation may be that traditional SDT approaches either incorporate lit tle or no physiology or make simplifying assumptions regarding the den sity functions describing the physiological data. In the work presente d here, an integrated approach, which combines SDT and a physiological ly based model of the human auditory system, is proposed as an alterna te method of quantifying detection performance. To validate this appro ach, the predicted detection performance for a simultaneous masking ta sk is compared to predictions obtained from traditional methods and to experimental data. Additionally, the sensitivity of the integrated me thod is thoroughly investigated. The results suggest that by combining SDT with a physiologically based auditory model, thereby capitalizing on the strengths of each individual method, the previously observed d iscrepancies can be partially explained as the result of physical proc esses inherent in the auditory system rather than unspecified ''intern al noise'' and more accurate predictions of psychophysical behavior ca n be obtained. (C) 1998 Acoustical Society of America.