Authors: Jiayue Liu *1 ; Josh Stohl 2; Tobias Overath 1
Affiliations:
1 Duke University
2 North American Research Laboratory, MED-EL Corporation
The past decade has seen numerous studies investigating the diagnosis, consequences, and possible treatments for cochlear synaptopathy. However, it remains unclear how the auditory cortex reacts to it. A recent mouse study found that, with 90% synapse loss, the auditory cortex shows hyper-synchronized neuronal activity (‘internal noise’) only in missed tone detection trials in noise (Resnik & Polley, 2021). This suggests that such cortical ‘internal noise’ might be responsible for degraded behavioral performance in noisy listening conditions. In this study, we investigated whether this ‘internal noise’ can also be captured with EEG in humans, and whether it correlates with indicators of hearing difficulties.
Twenty-three participants with near-normal hearing performed a monaural tone detection task in either quiet or noise while their EEG was recorded. Participants also performed tasks that have been suggested to reveal cochlear synaptopathy, including frequency modulation detection, the American English Matrix test in quiet and noise, ABR, pure tone audiometry, a lifetime noise exposure questionnaire and the Speech, Spatial and Quality of hearing questionnaire (SSQ). The analysis tested whether single trial EEG can predict behavior (hit vs. miss trials) and whether such EEG prediction correlates with other indicators of cochlear synaptopathy.
Ongoing EEG analyses suggest that pre-stimulus EEG activity does not predict behavioral outcomes. In contrast, significant prediction performance of post-stimulus EEG likely reflects the presence of the P300 component for hit trials, but not earlier auditory processing stages. This performance was correlated with age and SSQ, but not with any of the other measures.
Single-trial EEG decoding was successful on post-stimulus data, but at chance during the pre-stimulus period. EEG prediction performance was correlated with hearing difficulties in noise. More data is needed to determine this effect robustly.
