Date of Completion
speech perception, perceptual learning, auditory training, cochlear implants
Rachel M. Theodore
Emily B. Myers
James S. Magnuson
Field of Study
Speech, Language, and Hearing Sciences
Doctor of Philosophy
A challenge in the cochlear implant adaptation literature is accounting for wide variability in speech perception outcomes following implantation. While many preimplantation factors contribute to the variance observed in outcomes, formal auditory training has been proposed as a way to maximize speech comprehension benefits for cochlear implant users. However, there is currently no standardized set of recommendations as to how training should be implemented in the patient population. The goal of this dissertation was to examine two potential training variables of interest: time of day in which the training occurs and the task structure of training. This dissertation examined these factors in the context of speech learning for normal hearing listeners adapting to degraded speech input as a first step towards identifying training parameters that could hold clinical utility.
In chapter one, an interdisciplinary perspective is used to bridge the clinical rehabilitation literature with basic research examining perceptual learning of speech. Findings demonstrating a facilitative role of auditory training on adaptation to degraded speech signals in normal hearing listeners are reviewed in the context of rehabilitation for cochlear implant users. In chapter two, we examine the influence of sleep-based memory consolidation on perceptual learning of degraded speech, which has been previously shown to improve perception of other novel speech inputs. For listeners trained in close proximity to sleep, a significant stabilization effect was observed for trained and novel items that persisted at a one-week follow-up interval. This finding demonstrates a significant influence of sleep-based learning paradigms on perceptual learning for acoustically degraded speech signals. Finally, in chapter three, we examined the role of training task on perceptual learning outcomes both in the immediate and a later time interval. Across training groups, listeners demonstrated robust perceptual learning, regardless of whether the task was linguistic or non-linguistic and in the absence of linguistic feedback. This finding suggests alternate routes to degraded speech learning that may be customized based on the needs of the listener. Each section includes recommendations for future research that could foster translation of principles of speech learning in normal hearing listeners to aural rehabilitation protocols for cochlear implant patients.
Drouin, Julia R., "Perceptual Learning of Noise-Vocoded Speech" (2020). Doctoral Dissertations. 2588.