Dissertation
Cortical Dynamics of Speech-in-Noise Processing with Noise Reduction in Hearing Aids
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Spring 2020
DOI: 10.17077/etd.005294
Abstract
Noise reduction (NR) is a widely used signal processing scheme in modern hearing aids that aims to increase the ease and comfort of listening. The problem is that even the state-of-art NR algorithms of the current time induce inevitable distortion of speech cues while they attenuate background noise, which may have resulted in inconsistent benefits reported in prior studies. It is also challenging for audiologists to provide the best configurations for NR since each listener reacts differently to this trade-off between noise attenuation and speech-cue distortion. To provide a more systematic approach to selecting NR configurations for each hearing aid user, we need to understand the neurophysiology of suprathreshold processing of aided speech processing. Thus, this dissertation presents two aims. First, the present study investigated the neural mechanisms underlying the effect of NR on speech-in-noise recognition. We hypothesized that when the noise was attenuated using NR, cortical evoked responses would show the spatiotemporal pattern of efficient speech processing across the dorsal stream pathway despite inevitable spectral distortion. Second, this work identified factors that explain the individual differences in NR benefits. We hypothesized that different ability to extract speech signals out of background noise would drive individual differences in NR benefits. Here we showed that the NR condition, compared to the no-NR condition, invoked stronger early evoked response in the left supramarginal gyrus and weaker late evoked response in the left inferior frontal gyrus in the most challenging signal-to-noise ratio condition. This spatiotemporal pattern of cortical evoked responses induced by NR is comparable to the neural markers of efficient speech processing in relation to the change in noise level without NR reported from previous findings. We also found that listeners who had poor speech unmasking ability indexed by the amplitude ratio of the target word- and noise-evoked responses in the right Heschl’s gyrus got more benefits from NR. The present study provides mechanistic evidence of NR benefits, in addition to traditional outcome measures. This study also suggests many fundamental ideas for how to personalize the hearing aid configuration for NR based on listeners’ cortical activity, which will have a substantial clinical impact.
Details
- Title: Subtitle
- Cortical Dynamics of Speech-in-Noise Processing with Noise Reduction in Hearing Aids
- Creators
- Subong Kim
- Contributors
- Inyong Choi (Advisor)Yu-Hsiang Wu (Advisor)Carolyn J Brown (Committee Member)Shawn S Goodman (Committee Member)Camille C Dunn (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Speech and Hearing Science
- Date degree season
- Spring 2020
- DOI
- 10.17077/etd.005294
- Publisher
- University of Iowa
- Number of pages
- x, 79 pages
- Copyright
- Copyright 2020 Subong Kim
- Language
- English
- Description illustrations
- color illustrations
- Description bibliographic
- Includes bibliographical references (pages 71-78).
- Public Abstract (ETD)
Difficulty understanding speech in noise is one of the biggest complaints from hearing aid users. In the clinic, signal processing strategies in hearing aids are understood to suppress the noise and enhance speech signals for hearing aid users. Noise reduction is one of those features that is mostly applied in current hearing aids. However, it is well known that noise reduction does not help people to understand better speech in noise, whereas some cognitive benefits such as the increase in ease and comfort of listening are still reported. Thus, the present study investigated what actually happened in the brain, while participants were listening to speech stimuli with noise reduction at multiple listening conditions. Noise reduction turned out to facilitate more efficient speech-in-noise processing in the brain despite the fact that no benefit appeared in the behavioral scores on speech-in-noise tasks and subjective ratings. We also examined each individual’s different ability to extract speech signals from background noise based on their brain activity obtained at the auditory cortex. This ability may determine potential benefits expected from noise reduction usage in terms of behavioral performance and speech processing in the brain. The present study provides a novel way to measure the potential benefits of using noise reduction in hearing aids. Study results will facilitate personalized hearing aid fitting for noise reduction based on hearing aid users’ brain activity.
- Academic Unit
- Communication Sciences and Disorders
- Record Identifier
- 9983956195202771
Metrics
102 File views/ downloads
156 Record Views