ABSTRACT
This chapter explores two models of the type are applied to data from experiments on listeners with hearing impairments. The first model, developed by H. S. Colburn, R M. Zurek, K. G. Gabriel, and N. I. Durlach, uses a weighted linear combination to generate a decision variable that is analogous to lateral position. The second model assumes an optimum combination of interaural time and intensity differences, which for most normal-hearing listeners results in predicted binaural detection performance that is better than observed. The fundamental assumption is that interaural differences in the stimulus are estimated by interaural processors that operate separately on outputs of peripheral filters applied to the stimulus waveforms. The optimum processing of interaural differences predicts performance equal to or better than that observed in binaural detection experiments. Although binaural phenomena have generated a considerable number of mathematical and computational models, there have been few attempts to quantitatively model the effects of hearing impairments on binaural processing.
