Auditory backward masking and the effect of training in normal hearing adults.

Abstract

Recent renewed interest in auditory backward masking has stemmed from studies of children with language impairments who were found to have significant, elevated thresholds for this paradigm, compared to normal cohorts. There are, however, many unresolved theoretical and procedural issues. The present study was conducted to investigate some of these issues. Specific purposes were: 1) To establish the differences between backward masking and simultaneous masking in normal hearing subjects; 2) To investigate the effect of listening conditions, i.e., monaural versus binaural; 3) To measure the effect of training on the performance of a backward masking paradigm; and 4) To measure generalization of the trained condition to untrained conditions. Two experiments were conducted: In experiment I, we compared the performance of 10 normal-hearing adult subjects for backward masking and simultaneous masking paradigms in both monaural and binaural modes of presentation. Stimuli consisted of a 1000 Hz pure-tone and bandpass masker (600-1400 Hz). Tone thresholds for backward masking were significantly lower than those of the simultaneous masking. The binaural mode of presentation yielded lower thresholds than those of the monaural mode only for the backward masking condition. A significantly large inter- and intra-subject variability was observed in the backward masking paradigm. Experiment II was conducted to measure the effects of training and generalization. Two groups of subjects were included: a trained group and a control group of six and eight normal hearing adults, respectively. The trained group received 10 sessions of backward masking training with feedback. The control group was tested only twice, with a 2-week interval between testing. Only the trained group showed a significant improvement in backward masking thresholds, which amounted to an average of 10.7 dB. No significant improvement was observed in the non-trained group. A nonsignificant trend of generalization occurred for the trained task to the untrained ear. No generalization was evident in the untrained condition (simultaneous masking). The data have important clinical and theoretical implications regarding the ability to train for auditory tasks in general, and for backward masking in particular.