Purpose: The generalization of non-linguistic auditory perceptual training to syllable discrimination was investigated in two experiments.
Methods: Participants were divided into a control and training group. Both groups came for pre and post-testing sessions spaced ten days apart. Following pre- testing, the training group also participated for five consecutive days in non-linguistic auditory perceptual training. Training was adaptive and involved active sequencing of rising and falling frequency modulated sweeps for 30 minutes per day. Sweeps were passively varied in onset frequency, duration and rate of presentation. A syllable discrimination threshold (SDT) task was used as the pre and post-test measure. In experiment 1, a /ba/-/da/ syllable continuum was used. In experiment 2, the pre-test battery was expanded to include /ba/-/da/, /ba/-/wa/, and /sa/-/sta/ syllable continua and a tone sequencing task that mimicked other parameters of training.
Results: Results of experiment 1 revealed that the training group had a significantly lowered (better) SDT following training as compared to the control group. The extent of training-driven perceptual gain was significantly correlated with pre-training performance. In experiment 2, training resulted in a significantly lowered SDT for /ba/-/da/, but not for the other syllables or the tone sequencing task.
Conclusions: Results showed that task-specific attention drives generalization of auditory perceptual training from non-linguistic to linguistic contexts. Furthermore, individual differences in initial perceptual performance affect the degree of generalization following training.