Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Developmental Patterns of Brain–Speech Synchrony in Mandarin-Speaking Children: An EEG-TRF Study
Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai
Yuchun Chen1, Chun-Hsien Hsu2, Hong-Hsiang Liu1; 1Fu Jen Catholic University, 2National Central University
Brain-to-speech synchrony has been considered as an important neural mechanism supporting speech perception because it reflects how the brain responses to speech stimuli temporally aligns with acoustic and linguistic information in continuous speech. Recent evidence suggests that stronger neural envelope tracking is associated with better speech intelligibility in children and second language learners. This raises the possibility that brain-to-speech synchrony reflects not only low-level acoustic tracking, but also the perceptual accessibility and linguistic interpretability of continuous speech. The present study aims to investigate the developmental pattern of brain-to-speech synchrony in Mandarin-speaking children by comparing temporal response functions elicited by normal and reversed speech. Because reversed speech preserves many acoustic envelope properties while disrupting linguistic interpretation, this comparison allows us to examine whether cortical tracking is enhanced for intelligible speech beyond acoustic envelope processing alone. Fourteen typically developing children aged 4 to 8 years participated in a speech-listening experiment while EEG was recorded. Participants listened to age-appropriate stories for comprehension and also listened to reversed-speech stimuli. Two stories were presented in total, with story order counterbalanced across participants. Brain-to-speech synchrony was quantified using temporal response function (TRF) analysis. Specifically, speech envelope and envelope-onset predictors were extracted from each stimulus using a gammatone filterbank, and both EEG and acoustic predictors were resampled to 100 Hz and aligned using event markers. TRFs were estimated separately for each participant and speech condition, including normal and reversed speech, using a boosting algorithm (implemented in the Eelbrain toolbox) with a response window of −100 to 500 ms. A cluster-based permutation test was used to control for multiple comparisons and to identify spatiotemporal clusters in which TRF amplitudes differed significantly between the normal and reversed speech conditions. For the speech-envelope TRF, a significant spatiotemporal cluster was observed over fronto-central scalp sites between 210 and 390 ms, with greater amplitudes for reversed speech than for normal speech, suggesting enhanced tracking of global envelope fluctuations or increased sensory processing of acoustically unusual, unintelligible input. In contrast, the envelope-onset TRF showed a significant spatiotemporal cluster over the left central scalp between 170 and 560 ms, with greater amplitudes for normal speech than for reversed speech (Normal > Reversed). Because envelope onset cues are closely related to speech-event boundaries and temporal segmentation, this pattern suggests that intelligible speech may enhance children’s cortical sensitivity to acoustic landmarks that support speech comprehension. In conclusion, these preliminary findings from typically developing children suggest that brain-to-speech synchrony is sensitive to the distinction between intelligible and unintelligible continuous speech. Rather than reflecting a single acoustic-tracking mechanism, TRF responses appear to depend on the speech feature being modeled, with envelope and envelope-onset cues indexing partly distinct aspects of cortical speech processing. These results suggest that envelope-onset tracking may be especially relevant for characterizing neural mechanisms that support speech segmentation and comprehension during typical development.
Topic Areas: Language Development/Acquisition,