5aPP8. Learning and generalization in auditory backward masking.

Session: Friday Morning, December 6

Time: 9:45


Author: Beverly A. Wright
Location: Keck Ctr., Box 0732, Univ. of California, San Francisco, CA 94143-0732
Author: Paul A. Johnston
Location: Keck Ctr., Box 0732, Univ. of California, San Francisco, CA 94143-0732
Author: Miriam D. Reid
Location: Keck Ctr., Box 0732, Univ. of California, San Francisco, CA 94143-0732

Abstract:

Excessive auditory backward masking has been observed in elderly persons and in children with language impairments. The purposes of this project were to determine (1) whether practice can improve the ability to detect a backward-masked signal and, if so, (2) whether this improvement generalizes to other masking tasks. These issues were examined in six normal-hearing adults using an adaptive, 2IFC procedure. Each subject was asked to detect a tonal signal (10 ms, 1000 Hz) presented immediately before a noise masker (300 ms, 200--1800 Hz, 40-dB SPL spectrum level), on 900 trials per day for ten days. Before and after the training period, subjects were tested on five other masking tasks in addition to the trained task. A one-way ANOVA revealed a significant learning curve over the ten days of training [F(9,45)=5.1, p<0.001], with the average signal threshold decreasing from 63 to 53 dB SPL. Comparison of the pre- and post-test results showed significant threshold reductions in other backward-masking tasks, but not in simultaneous- and forward-masking tasks. These results indicate that (1) the interference from an auditory backward masker can be reduced with practice, and that (2) this learning is specific to backward masking. [Work supported by the James S. McDonnell Foundation and NIDCD.]


ASA 132nd meeting - Hawaii, December 1996