Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
The functional connectome and neural tracking of natural speech in post-stroke aphasia
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster E51 in Poster Session E, Thursday, October 26, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Also presenting in Lightning Talks E, Thursday, October 26, 10:00 - 10:15 am CEST, Auditorium
Ramtin Mehraram1, Jill Kries1, Pieter De Clercq1, Tom Francart1, Maaike Vandermosten1; 1KU Leuven
Aphasia is an impairment of language processing most commonly caused by a stroke. Using EEG, we recently detected significant alteration of functional network properties in persons with chronic post-stroke aphasia (PWA) as compared with healthy controls (HC) during natural speech listening [1]. Furthermore, we reported between-group differences in neural tracking of natural speech ([2], [3]). Here we combine the two approaches to assess the functional connectome associated with neural tracking of natural speech and its alteration in PWA. Our participant cohort comprised 23 HC and 43 PWA. EEG (64 channel) was recorded while the participants listened to a 25-minute-long story. After undergoing pre-processing, the recordings were filtered within theta (4.5-7 Hz) and low-gamma (31-49 Hz) frequency bands, where a functional network disruption emerged for the EEG time series [1]. Neural tracking of the speech envelope from each EEG signals was estimated as temporal mutual information function (TMIF) [4], and connectivity between TMIFs was computed as weighted phase lag index. Significant within-group and differential between-group network patterns were assessed using the Network-Based Statistics (NBS) toolbox. Functional network properties were investigated by means of graph theory. Accuracy of EEG-based network metrics to classify individuals as HC or PWA was explored by means of a random forest classifier. The NBS revealed a significant TMIF-network component in both theta- and low-gamma-bands (P < 0.001). However, only the latter showed an alteration in PWA, namely a weaker cluster connecting frontal with left and right temporo-occipital scalp regions (P < 0.005). Neural tracking in PWA exhibited a more pronounced small-world network compared to HC, together with higher network segregation. Altogether, TMIF-network metrics could correctly classify the participants with area under the receiver operating characteristic (ROC) curve of 74%. Our combined analysis proved that neural tracking of natural speech across different brain regions occurs in a functionally connected manner, and that this process is affected in PWA. Interestingly, only the low-gamma-band exhibited altered between-region TMIF synchronization in PWA, despite differences between groups were previously detected also in the theta-band for both connectivity of EEG time-series [1] and neural tracking [3]. Overall, these novel findings further contribute to disentangling the neurobiological mechanisms associated with language processing in aphasia. References: [1] R. Mehraram, J. Kries, P. De Clercq, M. Vandermosten, and T. Francart, ‘EEG reveals brain network alterations in chronic aphasia during natural speech listening’, bioRxiv, p. 2023.03.10.532034, Mar. 2023, doi: 10.1101/2023.03.10.532034. [2] J. Kries et al., ‘Exploring neural tracking of acoustic and linguistic speech representations in individuals with post-stroke aphasia’, bioRxiv, p. 2023.03.01.530707, Mar. 2023, doi: 10.1101/2023.03.01.530707. [3] P. De Clercq, J. Kries, R. Mehraram, J. Vanthornhout, T. Francart, and M. Vandermosten, ‘Detecting post-stroke aphasia using EEG-based neural envelope tracking of natural speech’, medRxiv, p. 2023.03.14.23287194, Mar. 2023, doi: 10.1101/2023.03.14.23287194. [4] P. De Clercq, J. Vanthornhout, M. Vandermosten, and T. Francart, ‘Beyond linear neural envelope tracking: a mutual information approach’, J Neural Eng, vol. 20, no. 2, p. 026007, Mar. 2023, doi: 10.1088/1741-2552/ACBE1D.
Topic Areas: Disorders: Acquired, Speech Perception