ORCID

http://orcid.org/0000-0003-1054-5701

Date of Award

Spring 2017

Document Type

Dissertation

Degree Name

Doctor of Audiology (AuD)

Department

Department of Communication Sciences and Disorders

Advisor(s)

Lincoln C. Gray

Bradley W. Kesse

Robert L. Nagel

Christopher G. Clinard

Abstract

MANUSCRIPT #1 TITLE

Development of a Deployable Stereo-Hearing System

ABSTRACT

Objective: To investigate the effectiveness and efficiency of a portable stereo-hearing testing system with the intent of deployment for data collection in future studies. We quantify sound localization accuracy and speech-in-noise thresholds comparing unilateral (such as single-sided deafness) and bilateral subjects. We desired to design a small, inexpensive system that would show a large effect size between binaural and monaural subjects in a variety of stereo hearing tasks.

Methods: Subjects were tested on localization accuracy and speech understanding in noise using a laboratory-made stereo-hearing testing device. For the localization task, the subject identifies the location of a 250 ms noise presented randomly at one of 8 speakers in a 180o array. For the speech-in-noise task, the subject identifies the CRM color/number command presented from a speaker in one hemi-field while an adaptive noise track is presented simultaneously from a speaker in the opposite hemi-field. RMS error quantified accuracy of localization. Analysis of the speech-detection thresholds involved differences between conditions when the speech and noise were each on different sides: a best-to-worst condition compared speech toward the good ear and noise toward the bad ear and vice versa; in a good-to-poor condition both the speech and noise were closer to midline (approaching straight ahead of the subject) but still on opposite hemi-fields. The right ear was considered ‘good’ for bilaterally-normal control subjects. We expect a large difference between the ‘best’ and ‘worst’ conditions for unilateral subjects and no such difference for the bilateral controls.

Results: Differences exist between unilateral and bilateral subjects in both sound localization and understanding speech in noise with ‘extremely large’ effect sizes (Cohen’s d >3 or ‘huge’ for both tests (and 1.6 or twice what Cohen said was a ‘large effect’ for the ‘good-to-poor’ comparison). Bilateral subjects localized and listened in noise better than unilateral subjects (p

Conclusions: Our device can distinguish between monaural and binaural subjects and is ready for deployment to investigate patients after otologic surgery, who we expect to perform ‘between’ these unilateral and bilateral subjects.Bilateral subjects localized sounds with near-perfect accuracy while unilateral subjects made many more errors. In unilateral subjects, thresholds detecting signals in noise were highly dependent on which hemi-fields produced the signal and noise (as expected, participants with unilateral hearing heard better when the signal was on the side of their one normal ear with contralateral noise facing the poorer ear) while bilateral subjects were relatively unaffected by the hemi-field of signal and noise. Plugged subjects (bilateral controls with an ear plug and an earmuff) localized worse than true unilateral subjects, suggesting an effect of learning how to localize with monaural cues.



MANUSCRIPT #2 TITLE

Deployment of a Stereo-Hearing System to Postoperative Atresia Patients

ABSTRACT

Objective: To investigate the performance of atresia patients, postoperatively, in two binaural processing tasks, and to compare that performance to subjects with normal hearing and subjects with a complete unilateral hearing loss. To investigate the reliability, validity, and efficiency of collecting data via transcontinental shipment of the prototype device.

Methods: From their home, subjects were tested on localization accuracy and speech understanding in noise using a laboratory-made stereo-hearing testing device. For the localization task, the subject identified the location of a 250 ms noise burst presented randomly from an 8-speaker array. For the speech-in-noise task, the subject must identify a color/number command presented from a speaker in one hemi-field while an adaptive noise track is presented simultaneously from a speaker in the adjacent hemi-field. The test is repeated with different locations of speech and noise.

Results: Postoperative atresia subjects performed better than unilateral subjects in all tasks, showing the expected improvements in binaural processing following canal-plasty. Atresia patients equaled the bilateral controls in sound localization and the most challenging speech in noise test. Within this initial sample of atresia patients (N=9) only vague trends are evident among the dependent variables and with various covariates. Post-operative audiometric speech reception thresholds correlate with the speech in noise testing. Both localization and speech-in-noise understanding appear to improve over post-operative time as a general trend, but outlier(s) increase the variance to a point of statistical insignificance.

Conclusions: Repaired atresia subjects perform better than unilateral subjects in localizing sound and understanding speech with a separated noise. The performance of the post-op atresia patients is closer to the bilateral subjects than to unilateral subjects, confirming a benefit of the surgical repair of congenital conductive hearing loss. Our device can be used reliably and efficiently to collect data via transcontinental shipment to residential locations. More follow-up longitudinal data would help investigate any improvement in of binaural listening tasks over time in these patients. One option to continue this research is for medical centers to deploy this device to patients’ homes annually for updated testing.

Available for download on Wednesday, April 10, 2019

Share

COinS