Data-driven spatial filtering for improved measurement of cortical tracking of multiple representations of speech

Abstract

ObjectiveMeasurement of the cortical tracking of continuous natural speech from electroencephalography (EEG) recordings using a forward model is becoming an important tool in auditory neuroscience. However, it requires a manual channel selection based on visual inspection or prior knowledge to obtain a summary measure of cortical tracking. In this study, we present a method to on the one hand remove non-stimulus-related activity from the EEG signals to be predicted, and on the other hand automatically select the channels of interest. We also aim to show that the EEG prediction from phonology-related speech features is possible in Dutch. ApproachEighteen participants listened to a Flemish story, while their EEG was recorded. Subject-specific and grand-average temporal responses functions were determined between the EEG activity in different frequency bands and several stimulus features: the envelope, spectrogram, phonemes, phonetic features or a combination. The temporal response functions were then used to predict EEG from the stimulus, and the predicted was compared with the recorded EEG, yielding a measure of cortical tracking of stimulus features. A spatial filter was calculated based on the generalized eigenvalue decomposition (GEVD), and the effect on EEG prediction accuracy was determined. Main resultsA model including both low- and high-level speech representations was able to better predict the brain responses to the speech than a model only including low-level features. The inclusion of a GEVD-based spatial filter in the model increased the prediction accuracy of cortical responses to each speech feature at both single-subject (270% improvement) and group-level (310 %). SignificanceWe showed that the inclusion of acoustical and phonetic speech information and the addition of a data-driven spatial filter allow improved modelling of the relationship between the speech and its brain response and offer an automatic channel selection. HighlightsO_LIAutomatic channel selection for evaluating the cortical tracking of continuous natural speech C_LIO_LIData-driven spatial filtering for removing non-stimulus-related activity from the EEG signals C_LIO_LIImproved prediction of brain responses to speech by combining acoustical and phonetic speech information in Dutch C_LI DisclosureThe authors report no disclosures relevant to the manuscript.