Age-related differences in the sequential organization of speech sounds

Abstract

This study investigated the effects of age on listeners' tendency to group speech tokens into one or two auditory streams. Younger and older adults were presented with sequences of four vowel sounds, which were arranged according to the proximity of first-formant frequencies between adjacent vowels. In Experiment 1, participants were less accurate in identifying the order of the four vowels and more likely to report hearing two streams when the first-formant alternated between low and high frequency and the overall difference between adjacent vowels was large. This effect of first-formant continuity on temporal order judgments and probability of hearing two streams was higher in younger than in older adults. In Experiment 2, participants indicated whether there was rhythm irregularity in an otherwise isochronous sequence of four vowels. Young adults' thresholds were lower when successive first-formants ascended or descended monotonically (condition promoting integration) than when they alternated discontinuously (condition promoting streaming). This effect was not observed in older adults whose thresholds were comparable for both types of vowel sequences. These two experiments provide converging evidence for an age-related deficit in exploiting first-formant information between consecutive vowels, which appear to impede older adults' ability to sequentially group speech sounds over time.