Enhancing older adult speech perception in challenging listening environments : contextual cues and music training




Smayda, Kirsten Elisabeth

Journal Title

Journal ISSN

Volume Title



Normal aging is associated with difficulty understanding speech in adverse listening conditions and can lead to problems for the elderly such as social isolation, anxiety, depression, and diminished quality of life. A large literature suggests at least two types of noise can negatively interfere with speech intelligibility: energetic and informational noise. Energetic masking results when the noise spectro-temporally overlaps with the speech signal (e.g., near a construction site). Informational masking results when the noise contains information beyond spectro-temporal overlap with the speech signal (e.g., the ‘cocktail’ party situation). Cognitive processes have been implicated in mediating individual differences in speech-in-noise (SPIN) perception such as auditory working memory, attention, and processing speed; as well as perceptual processes such as temporal processing and gap detection. Importantly, the cognitive and perceptual subprocesses involved in accurate speech-in-noise perception also decline as we age. An expansive literature suggests that music training is positively associated with enhancements in not only SPIN processing, but also the perceptual and cognitive abilities supporting SPIN perception. Importantly, the causal effect of music training on older adult SPIN perception is poorly understood. The overarching goal of this thesis is to characterize the contextual and listener features that can improve older adult speech-in-noise perception. The first paper in this dissertation explores the extent to which contextual cues, such as visual and semantic information, can aid in older adult speech-in-noise processing. In Paper 2, we examine the source of a musician advantage in learning novel speech categories. Using computational modeling we show that the musician advantage is due to both cognitive and perceptual processes. Paper 3 tests the extent to which age of onset of music training improves decision-making later in life. The broader implications of Papers 1 through 3 are explored in the General Discussion, which includes a proof-of-concept training study experimentally testing the effect of ten weeks of group piano lessons on older adult speech-in-noise processing. Preliminary results suggest that music training confers larger SPIN improvements relative to no training, and participants in the music training condition were more motivated to complete their training relative to those in the active control group.



LCSH Subject Headings