Design of a distributable stereo hearing test package

Abstract

The localization and identification of sounds in background noise are such important auditory processing skills that a significant amount of incompetency may lead to various confusions and learning delays. Through a partnership with James Madison University and the University of Virginia, a unique opportunity exists to test patients before and after a corrected maximal conductive hearing loss in one ear. Patients with congenital aural atresia come to the University of Virginia for surgery that will give them normal hearing. Insurance pays for a pure tone threshold hearing test one month after surgery, but due to cost restrictions, longitudinal follow-up testing is often not performed. However, longitudinal data from follow-up studies is essential for understanding the effectiveness of the surgery. This paper focuses on the design, construction, and testing of a prototype shippable hearing test system for client use and patient testing. The system tests two binaural hearing abilities-the ability to isolate a spatially separated signal from noise and the ability to localize the source of a sound. This work has been performed as a partnership between the Department of Engineering and the Department of Communication Sciences and Disorders at James Madison University. The student team consists of two juniors and four seniors working on their two-year engineering capstone project in the Department of Engineering at James Madison University, as well as a senior studying Audiology & Speech Pathology working on her Honors thesis. This project has followed an engineering design process beginning with collecting customer needs and understanding required specifications and ending with prototype testing and refinement. Validation testing of the prototype testing system was performed with listeners with both normal hearing and with unilateral hearing loss. In initial testing, the RMS localization errors were measured from 19 control listeners with normal hearing and four with complete unilateral hearing loss. There was a significant difference between those listeners with one good ear versus those with two good ears (p=.01, Cohen's D > 1 or `large' effect size). These results provide promise as to the effectiveness of the designed testing package.