Neural Speech Tracking with EEG: Integrating Acoustics and Linguistics for Hearing Aid Users

Area of Science:

• Neuroscience
• Audiology
• Computational Linguistics

Background:

• Understanding speech in noisy environments is challenging for hearing aid users.
• Neural processing of speech features in hearing aid users requires further investigation.
• The impact of hearing aid noise reduction on neural speech tracking is not well understood.

Purpose of the Study:

• To investigate the neural encoding of speech features in hearing aid users.
• To examine how hearing aid noise reduction affects neural tracking of acoustic and linguistic speech features.
• To correlate specific speech features with electroencephalography (EEG) signals during a simulated cocktail party.

Main Methods:

• Utilized electroencephalography (EEG) in hearing aid users during a simulated cocktail party.
• Analyzed neural tracking of acoustic envelope, phonetic features, word onset, and word surprisal.
• Employed Temporal Response Functions (TRFs) with a boosting algorithm to correlate speech features with EEG signals.

Main Results:

• Significant distinctions were found in neural tracking between attended (target) and unattended (masker) speech.
• The acoustic envelope demonstrated the strongest correlation with EEG responses.
• Acoustic envelope and phonetic features correlated with early neural processing, while word onset and surprisal linked to later stages. Noise reduction modulated feature tracking.

Conclusions:

• Neural processing of speech in hearing aid users involves distinct tracking of various acoustic and linguistic features.
• Hearing aid noise reduction impacts the neural tracking of speech.
• Findings offer insights for developing adaptive hearing aids tailored to individual neural speech processing.