PP314 - Feasibility of a Virtual Environment to Assess Listening Outcomes

Abstract:
This research tested the feasibility of a new measure that enables clinicians to simultaneously assess speech understanding in noise and localization accuracy. The study was conducted in an anechoic chamber with an array of speakers located in the front hemifield. Participants wore a virtual reality headset and listened to speech stimuli in background noise. They completed two tasks simultaneously: localization and speech understanding. Participants selected the avatar that the speech was coming from. The intensity of the competing noise varied to find the level at which the participant could just follow the conversation.

Summary:
Rationale/Purpose
This research developed, piloted, and tested the feasibility of a new measure of speech communication in noise that enabled us to simultaneously assess speech understanding and localization accuracy.

Methods
The present study contained two phases. The first phase involved developing and piloting the measure on 4 young adults with typical hearing. The second phase involved validating the measure on older adults (8 with hearing loss and 8 with age-adjusted “normal” hearing). This second phase explores whether the measure is feasible for use with older adults, and whether it is sensitive to detecting differences in listening performance based on hearing ability.

The study took place in an anechoic chamber equipped with a truss containing a 360 degree array of speakers for the localization task, of which the front hemifield was utilized. Participants sat in a chair in the middle of the speaker array, with the speakers at ear level. A virtual reality environment was designed for the purpose of this study. Participants wore a virtual reality headset for the duration of the study, through which they saw avatars standing in the place of each speaker used for the speech stimuli (9 speakers in the front hemifield). Participants listened to speech stimuli, which were recorded by several different speakers, with each passage centered around a focal topic (i.e. windows, donkeys, gloves, etc) in the presence of background multi-talker babble. The Connected Speech Test was used. Participants completed two tasks simultaneously: localization and speech understanding. The speech stimulus switched between the 9 speakers in the front hemifield periodically, and participants were tasked with selecting the avatar with a hand-held controller to indicate where that the speech was coming from each time that it switched. From this task, both degree of error and reaction time to the speech stimulus were measured. Next, the intensity of the competing noise stimulus was varied throughout the experiment to find the signal to noise ratio at which the participant could just follow the conversation (JFC). This was accomplished by lowering the competing background noise each time the speakers switched until the participant indicated they could just follow the conversation by repeating the overall topic of the passage.

Results and Conclusions
Results supported the feasibility of using a virtual environment to assess speech understanding and localization accuracy for older adults. Pilot data demonstrated measurable differences between participant groups and listening conditions in expected directions, indicating sufficient sensitivity for these measures to detect existing differences in listening-related difficulty. Older adults with hearing loss required greater SNRs to just follow the conversation, had poorer localization accuracy, and longer localization response times compared to older adults with age-adjusted normal hearing, and both groups had poorer performance compared to young, listeners with typical hearing. In addition, localization accuracy and response times were poorer in the worst SNR conditions; however, due to lack of power these effects did not reach statistical significance.

Instructional Level
An intermediate instructional level was selected. This study focuses on a new measure of speech understanding and localization.