4aPP4. Comparisons between tone-in-noise and phase sensitivity tasks.

Session: Thursday Morning, June 19


Author: Brian Branstetter
Location: Dept. of Cognit. Sci., Univ. of California, Irvine, CA 92612, bgberg@uci.edu
Author: Bruce G. Berg
Location: Dept. of Cognit. Sci., Univ. of California, Irvine, CA 92612, bgberg@uci.edu
Author: Curt Southworth
Location: Dept. of Cognit. Sci., Univ. of California, Irvine, CA 92612, bgberg@uci.edu
Author: Lisa Khuu
Location: Dept. of Cognit. Sci., Univ. of California, Irvine, CA 92612, bgberg@uci.edu

Abstract:

According to critical band theory, phase sensitivity as a function of stimulus bandwidth and detection thresholds for a tone masked by noise are both indicators of the frequency resolution of peripheral auditory filters. The relation between these two measures is examined. For the phase sensitivity task, stimuli consist of three tones with constant intensity and randomly sampled phases. Two identical sounds are played on each trial together with a third having different randomly sampled phases. Listeners identify the odd sound. The frequency separation between tones is varied according to an adaptive staircase procedure which yields an estimate of the bandwidth at which the odd sound is identified with 71% accuracy. Estimates are obtained for center frequencies of 500, 1000, 2000, and 4000 Hz. For the second task, thresholds for 500, 1000, 2000, and 4000 Hz tones masked by broadband Gaussian noise are estimated. Results show that performance levels for different conditions within each task are highly correlated, whereas across-task correlations are either zero or slightly negative. These finding demonstrate that predictions from critical band theory do not generalize to the level of individual listeners. [Work supported by ONR.]


ASA 133rd meeting - Penn State, June 1997