Concepedia

TLDR

The study develops a quantitative black‑box model to interpret binaural masking‑level difference data. The model equalizes masking components across ears, subtracts them while accounting for error corruption, and is applied to pulsed‑tone signals masked by broadband Gaussian noise with only interaural time or amplitude differences. Results show the auditory system can adjust interaural intensity ratios and time delays only to about 1 dB and 150 µs, and struggles with delays exceeding the head‑width travel time.

Abstract

In this paper, a quantitative “black-box” model is developed for use in interpreting certain data on binaural-masking-level differences. The basic idea of this model is that the auditory system attempts to eliminate the masking components by first transforming the stimuli presented to the two ears so as to equalize the two masking components, and then subtracting. In order to obtain results that are quantitatively realistic, this processing is assumed to be corrupted by various types of errors. The model is applied to data in which the signal to be detected consists of a pulsed tone, the masking signal consists of loud, broad-band, Gaussian noise, and the only differences between the stimuli presented to the two ears are those of time delay or amplitude (the interaural amplitude ratios being restricted to unity or zero). The results indicate that the ability of the auditory. system to control interaural intensity ratios and interaural time delays is limited to accuracies of about 1 dB and 150 μsec and that the auditory system has difficulty in compensating for interaural time delays greater than the time it takes for sound to travel a distance equal to the width of the head.