Auditory integration training: Its effect on the perception of English of native and nonnative speakers

Abstract

This study examines the efficacy of Auditory Integration Training (AIT) as a means of perceptual training for adult native and nonnative speakers of English. AIT differs from other methods of auditory training that use either synthetic or human speech as the training stimulus, by using music as its stimulus. During AIT, the music can be altered in two ways. First, optional filters can be selected to eliminate specific frequencies. Second the music is always modulated at random by 20 dB during AIT. The effect of the filtering is that the music sounds less than clear and, from the modulation, the volume increases and decreases very quickly. Currently, this training method is used for children with a variety of language impairments, including hypersensitive hearing and learning disabilities. Developed by Berard (1993), it is considered an exercise of the perceptual system by its proponents who report improvement following AIT in both perception and production. The 43 participants in this study were native speakers of English, Japanese, and Korean and included both students and professors at the University of Arizona. The effect of perceptual training was evaluated by comparing performance in a pretest and posttest administered before and after AIT. Four tests were used: (1) Speech Perception in Noise (SPIN), (2) R-L word identification, (3) Synthetic /ra/-/la/ consonant discrimination, and (4) Synthetic /e/-/epsilon/ vowel discrimination. AIT was administered for one hour per day for 10 days. The first experiment examined the effect of AIT on monolingual English speakers. The results from the first experiment revealed significant improvement in the SPIN test, and changes in the synthetic /ra/-/la/ and /e/-/epsilon/ syllables. The second experiment examined the effect of AIT on native Japanese and Korean speakers and found one significant change---the manner in which native Japanese speakers categorized [e]. However, a control group of native English speakers who listened to unaltered music revealed improvements in the SPIN test similar to those in the first experiment. The improvements from AIT appear to be from the music itself and not from any filtering or modulation.