Binaural masking-level differences for interaurally delayed and for jittered pulse-train signals

Abstract

Two experiments were conducted to investigate the effects of interaural signal delay and temporal jitter on the MLD for pulse-train signals. A two-interval forced-choice detection task was employed with the level of the masking noise (No) controlled by a PEST program. In all conditions the signal in one ear was a pulse train composed of 100-μsec positive or negative polarity pulses with an interpulse period of 2 msec. The signal in the opposite ear was either an identical but temporally delayed pulse train (Experiment 1) or was a pulse train in which the interpulse interval was perturbed (jittered) for some proportion of the pulses (Experiment 2). In both experiments conditions were run in which pulse polarity was either fixed or was a random variable. All signals were low-pass filtered at 5 kHz. In Experiment 1 the magnitude of the MLD was found to be a function of the interaural signal delay and the polarity pattern. In Experiment 2 the MLD was found to be a function of both the amount of temporal displacement and the proportion of pulses which were jittered. Results are interpreted with respect to the spectral properties of the signals. Relationships to previous MLD research and implications for correlation-type MLD models are discussed.