Previously we reported that pulse triplets with both interpulse intervals (IPI) equal to 2.0 ms could be discriminated from pulse triplets with each IPI differing from 2.0 ms by at least 0.14 ms (M. G. Ceruti, R. W. Floyd, and D. W. Martin, J. Acoust. Soc. Am. Suppl. 1 72, S92 (1982)]. Here we report a further experiment in which we presented subjects with a triple pulse stimulus with random differences in IPI's. The Objective of this experiment was to determine the accuracy with which subjects could manipulate the timing of a variable pulse, relative to a pulse doublet with a fixed 4.0-ms interval, in order to produce a triplet containing two IPI's equal to 2.0 ms. The time delay of the doublet, relative to the variable pulse, was randomized prior to each trial. All three identical broadband pulses had durations of 1.0 ms each. Triple pulse stimuli were repeated at 9 Hz during each trial. The subjects' task was to listen to the TSP generated by each stimulus, to adjust a potentiometer controlling the timing of the variable pulse, and finally, to respond when the pulse was centered in the 4.0-ms pulse doublet. The data indicate that subjects responded correctly (forming two IPI's of exactly 2.0 ms each) on 41% of the trials, and produce intervals which both differed from 2.0 ms by 0.1 ms on 33% of the trials. We will discuss both the subjects' performance strategies for pulse centering, as well as subjects' pitch perception of the stimuli.
Read full abstract