Effects of Aging on Behavioral Measures of Listening Effort

Abstract

Difficulty with speech communication in noise is a common problem among elderly individuals. Older adults often report challenges with understanding speech, particularly in noisy environments. Growing evidence suggests that cognitive effort is a significant factor in speech understanding in noise. Although hearing loss is commonly experienced by older adults, according to prevalence estimates, about 4 in 10 adults age 65 and over will have impaired hearing. Older adults also experience decline in a number of cognitive abilities. The focus here was on aging alone to eliminate hearing loss as a contributing factor. The primary focus of this study was to measure cognitive effort (listening effort) in young and older adults with normal hearing while completing a speech in noise task. This study also examined some methodological issues for the measurement of listening effort. The most common means of behavioral assessment of listening effort is through use of a dual-task paradigm (DTP), whereby participants perform a “primary” speech-perception task along with a “secondary” task that does not involve speech perception. The two tasks can be administered concurrently or sequentially. It is not known whether DTPs administered sequentially and concurrently in the same person will yield similar results. The primary task in the DTP used here was a speech-identification task with a target talker and two competing talkers; the secondary task was either concurrent or sequential recall of a portion of the target message. Another methodological issue examined was the influence of the gender of the competing talker, either the same as (male) or different from (female) the target talker. The primary finding was that, when the performance of young and older adults was equated at baseline in the DTP, few effects of age on listening effort were seen. Differences between the concurrent and sequential conditions emerged, however, including a larger dual-task effect on the secondary task, slower response times, and poorer performance overall for the sequential condition. Consistent with previous findings in the literature, performance on the speech segregation portion of the DTP was generally better when the genders of the target and competing talkers differed.