Tracking Biomarkers of Speech Intelligibility

Difficulties understanding speech in noisy environments repeatedly coincide with high-frequency hearing loss. This complaint is commonly exhibited in adults in middle/older age who have a history of noise exposure. In this study, an immersive audiomotor training game will be utilized to drive improvements in speech intelligibility, controlled by an auditory memory training game. Physiological measures will be tracked that could inform clinical assessment of hearing in noise abilities.

Hearing loss is a chronic health disorder affecting approximately 15 percent of Americans. High-frequency hearing loss can be imperceptible when listening in quiet environments. However, in complex noisy environments, individuals can severely suffer from an inability to resolve speech. This is despite having normal audiometric thresholds in the low-frequency range within which speech signals are contained. There are several factors that could contribute to speech intelligibility difficulties that are imperceptible with typical hearing tests. These include impaired temporal encoding at the auditory nerve and the downstream sequelae of peripheral damage in the central auditory pathway. The goal of this study is to assess how a set of physiological measures of auditory/neural processing map onto suprathreshold hearing outcomes. In a previous study, significant improvements in speech intelligibility have resulted from training on an immersive video game. The game uses a closed-loop audiomotor interface design that reinforces sensory-guided feedback. As a control, an auditory memory training game has been developed to replicate user expectations and procedural learning. In this study, an audiomotor game and control game will be used as mechanisms through which to track changes to physiological and perceptual biomarkers in a high frequency hearing loss cohort and a matched normal hearing cohort. Tests will be run pre-training, post- training, and at a follow-up point.

Closed-loop audiomotor game. Home-based training sessions for 3.5 hours per week over an 8-week. Participants will have high frequency hearing loss.

Closed-loop audiomotor game. Home-based training sessions for 3.5 hours per week over an 8-week. Participants will have normal hearing.

Auditory memory game. Home-based training sessions for 3.5 hours per week over an 8-week. Participants will have high frequency hearing loss.

Auditory memory game. Home-based training sessions for 3.5 hours per week over an 8-week. Participants will have normal hearing.

Keyword recognition accuracy for sentences in noise will be assessed with a clinical test, the QuickSiN. Signal to noise ratios vary from 25 to 0 dB in 5 dB steps.

Growth in pupil dilation over the duration of the sentence will be measured while subjects perform the QuickSiN test at signal to noise ratios varying from 25 to 0 in 5 dB steps.

Inclusion Criteria: High frequency hearing loss subjects - Thresholds of 15 dB HL or less for 0.25-2 kHz. Thresholds of 30-60 dB HL from 3-8 kHz. Normal hearing subjects - Thresholds of 15 dB HL or less from 0.5-8 kHz. Thresholds between left and right ears symmetric within 10 dB for all frequencies. Native English speakers. Exclusion Criteria: Conductive hearing loss (as assessed by audiologist). Active otologic disease (as assessed by audiologist). Significant cognitive decline (Montreal Cognitive Assessment score > 25).