Hearing & Speech Sciences Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/2776
Browse
3 results
Search Results
Item INFLUENCE OF SUPPORTIVE CONTEXT AND STIMULUS VARIABILITY ON RAPID ADAPTATION TO NON-NATIVE SPEECH(2021) Bieber, Rebecca; Gordon-Salant, Sandra; Anderson, Samira; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Older listeners, particularly those with age-related hearing loss, report a high level of difficulty in perception of non-native speech when queried in clinical settings. In an increasingly global society, addressing these challenges is an important component of providing auditory care and rehabilitation to this population. Prior literature shows that younger listeners can quickly adapt to both unfamiliar and challenging auditory stimuli, improving their perception over a short period of exposure. Prior work has suggested that a protocol including higher variability of the speech materials may be most beneficial for learning; variability within the stimuli may serve to provide listeners with a larger range of acoustic information to map onto higher level lexical representations. However, there is also evidence that increased acoustic variability is not beneficial for all listeners. Listeners also benefit from the presence of semantic context during speech recognition tasks. It is less clear, however, whether older listeners derive more benefit than younger listeners from supportive context; some studies find increased benefit for older listeners, while others find that the context benefit is similar in magnitude across age groups.This project comprises a series of experiments utilizing behavioral and electrophysiologic measures designed to examine the contributions of acoustic variability and semantic context in relation to speech recognition during the course of rapid adaptation to non-native English speech. Experiment 1 examined the effects of increasing stimulus variability on behavioral measures of rapid adaptation. The results of the study indicated that stimulus variability impacted overall levels of recognition, but did not affect rate of adaptation. This was confirmed in Experiment 2, which also showed that degree of semantic context influenced rate of adaptation, but not overall performance levels. In Experiment 3, younger and older normal-hearing adults showed similar rates of adaptation to a non-native talker regardless of context level, though talker accent and context level interacted to the detriment of older listeners’ speech recognition. When cortical responses were examined, younger and older normal-hearing listeners showed similar predictive processing effects for both native and non-native speech.Item Effects of Age, Hearing Loss and Cognition on Discourse Comprehension and Speech Intelligibility Performance(2020) Schurman, Jaclyn; Gordon-Salant, Sandra; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Discourse comprehension requires listeners to interpret the meaning of an incoming message, integrate the message into memory and use the information to respond appropriately. Discourse comprehension is a skill required to effectively communicate with others in real time. The overall goal of this research is to determine the relative impact of multiple environmental and individual factors on discourse comprehension performance for younger and older adults with and without hearing loss using a clinically feasible testing approach. Study 1 focused on the impact of rapid speech on discourse comprehension performance for younger and older adults with and without hearing loss. Study 2 focused on the impact of background noise and masker type on discourse comprehension performance for younger and older adults with and without hearing loss. The influences of cognitive function and speech intelligibility were also of interest. The impact of these factors was measured using a self-selection paradigm in both studies. Listeners were required to self-select a time-compression ratio or signal-to-noise ratio (SNR) where they could understand and effectively answer questions about the discourse comprehension passages. Results showed that comprehension accuracy performance was held relatively constant across groups and conditions, but the time-compression ratios and SNRs varied significantly. Results in both studies demonstrated significant effects of age and hearing loss on the self-selection of listening rate and SNR. This result suggests that older adults are at a disadvantage for rapid speech and in the presence of background noise during a discourse comprehension task compared to younger adults. Older adults with hearing loss showed an additional disadvantage compared to older normal-hearing listeners for both difficult discourse comprehension tasks. Cognitive function, specifically processing speed and working memory, was shown to predict self-selected time-compression ratio and SNR. Understanding the effects of age, hearing loss and cognitive decline on discourse comprehension performance may eventually help mitigate these effects in real world listening situations.Item Effects of talker familiarity on speech understanding and cognitive effort in complex environments.(2020) Cohen, Julie; Gordon-Salant, Sandra; Brungart, Douglas S.; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The long-term goal of this project is to understand the cognitive mechanisms responsible for familiar voice (FV) benefit in real-world environments, and to develop means to exploit the FV benefit to increase saliency of attended speech for older adults with hearing loss. Older adults and those with hearing loss have greater difficulty in noisy environments than younger adults, due in part to a reduction in available cognitive resources. When older listeners are in a challenging environment, their reduced cognitive resources (i.e., working memory and inhibitory control) can result in increased listening effort to maintain speech understanding performance. Both younger and older listeners were tested in this study to determine if the familiar voice benefit varies with listener age under various listening conditions. Study 1 examined whether a FV improves speech understanding and working memory during a dynamic speech understanding task in a real-world setting for couples of younger and older adults. Results showed that both younger and older adults exhibited a talker familiarity benefit to speech understanding performance, but performance on a test of working memory capacity did not vary as a function of talker familiarity. Study 2 examined if a FV improves speech understanding in a simulated cocktail-party environment in a lab setting by presenting multi-talker stimuli that were either monotic or dichotic. Both YNH and ONH groups exhibited a familiarity benefit in monotic and dichotic listening conditions. However, results also showed that talker familiarity benefit in the monotic conditions varied as a function of talker identification accuracy. When the talker identification was correct, speech understanding was similar when listening to a familiar masker or when both voices were unfamiliar. However, when talker identification was incorrect, listening to a familiar masker resulted in a decline in speech understanding. Study 3 examined if a FV improves performance on a measure of auditory working memory. ONH listeners with higher working memory capacity exhibited a benefit in performance when listening to a familiar vs. unfamiliar target voice. Additionally, performance on the 1-back test varied as a function of working memory capacity and inhibitory control. Taken together, talker familiarity is a beneficial cue that both younger and older adults can utilize when listening in complex environments, such as a restaurant or a crowded gathering. Listening to a familiar voice can improve speech understanding in noise, particularly when the noise is composed of speech. However, this benefit did not impact performance on a high memory load task. Understanding the role that familiar voices may have on the allocation of cognitive resources could result in improved aural rehabilitation strategies and may ultimately facilitate improvements in partner communication in complex real-world environments.