Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Vocabulary Knowledge Modulates Neural Connectivity during Second Language Listening: An EEG Study
Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai
Takara Kenza Allal-Sumoto1, Hiroaki Mizuhara2; 1University of Hyogo, 2Nara Women's University
Vocabulary knowledge is a key factor in second language listening comprehension, yet how it shapes functional brain connectivity during online sentence processing remains unclear. This study examined whether differences in vocabulary knowledge levels; productive, only comprehensive, and noncomprehensive, are associated with changes in functional EEG connectivity and network organization during second language listening comprehension. Twenty-five Japanese learners of English participated in an EEG listening comprehension task using individualized stimulus sets based on their vocabulary knowledge profiles. Participants listened to English sentences of approximately ten words, each ending with a three- to four-word adverbial formulaic sequence, and rated their comprehension confidence after each sentence via button press. EEG was recorded with a 32-channel system. All analyses were time-locked to the onset of the final word of each formulaic sequence, with an analysis window spanning from 1000 ms before to 1500 ms after that onset. Phase-based functional connectivity was estimated using the debiased weighted phase lag index (wPLI), computed separately for theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz) bands across five 500 ms time windows relative to target word onset. Between-level connectivity differences were identified using the Network-Based Statistic (NBS). Within NBS-significant subnetworks, network organization was assessed via Newman-Girvan modularity, and node roles were characterized using betweenness centrality (BC), participation coefficient (P), and within-module degree z-score (Z-within). NBS revealed multiple frequency- and time-specific connectivity differences across vocabulary knowledge levels. Before target onset (−500 to 0 ms), an alpha-band subnetwork showed stronger connectivity for noncomprehensive relative to only comprehensive knowledge (p = 0.026; 40 edges). Immediately after target onset (0–500 ms), productive knowledge was associated with stronger connectivity in three subnetworks: theta-band (L2 < L3; p = 0.032; 9 edges), and two beta-band subnetworks (L1 < L3; p = 0.009; 33 edges; L2 < L3; p = 0.002; 39 edges). In the subsequent window (500–1000 ms), productive knowledge continued to show stronger connectivity in theta (L2 < L3; p = 0.036; 7 edges) and alpha bands (L1 < L3; p = 0.014; 31 edges). Taken together, these patterns may suggest that when lexical representations are less consolidated, greater top-down attentional and inhibitory control demands are reflected in pre-target alpha connectivity, while productive knowledge may support more efficient lexical access, semantic binding, and syntactic integration immediately after target onset, followed by sustained network coordination that could facilitate meaning consolidation in the later window. Node-role analyses revealed that connectivity increases associated with productive knowledge were accompanied by focal elevations in BC and P at fronto-central and occipital electrodes, indicating enhanced bridging-intermediary roles at these sites. In contrast, Z-within showed no consistent between-level differences, suggesting that productive knowledge modulates inter-regional information routing rather than within-module integration. These findings indicate that vocabulary knowledge shapes comprehension not only through local neural activation but also by altering how information is routed through functional networks. In particular, productive vocabulary knowledge was shown to enable more efficient processing not merely by accelerating processing speed, but through distinct patterns of functional brain connectivity.
Topic Areas: Language Development/Acquisition,