Quantitative Evaluation of a Lateralization Model of Masking-Level Differences

A lateralization model is proposed in which binaural detection in those cases that produce a masking-level difference is of a shift in the spatial locus of the auditory image. The shift is presumed to be a function of interaural time and intensity differences in the signal-plus-masker; it is quantified as a weighted sum of these two differences, averaged across the observation interval. The model is applied to a broad range of experimental data in which detection is given as a function of interaural phase and amplitude differences in the signal. It is shown here that, when applied to noise-masked low-frequency detection, the lateralization model is identical in its predictions to the time-difference model of Jeffress (1965) and the equalization-cancellation model of Durlach (1963). However, when applied to noise-masked high-frequency detection and to detection of signals masked by pure tones, the lateralization model better describes the data.