Poster Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions

Simultaneous interpreting transiently modulates neural signatures of working memory

Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai

Agustina Birba¹, Isabelle Chou², Guoqing Kwon³, Jiehui Hu³, Adolfo García^1,4,5; ¹Cognitive Neuroscience Center (CNC), Department of Life and Behavioral Sciences, University of San Andres, Buenos Aires, Argentina, ²College of Foreign Languages and Cultures, Sichuan University, Chengdu, Sichuan, China, ³School of Foreign Languages, University of Electronic Science and Technology of China, Chengdu, Sichuan, China, ⁴Global Brain Health Institute (GBHI), University of California San Francisco, United States; Trinity College Dublin (TCD), Dublin, Ireland, ⁵Departamento de Lingüística y Literatura, Facultad de Humanidades, Universidad de Santiago de Chile, Santiago, Chile

Simultaneous interpreting (SI) is one of the most cognitively demanding forms of bilingual language use, heavily taxing working memory (WM) processes. However, little is known about the immediate neurophysiological consequences of SI-related exertion on subsequent WM functioning. In the present study, we examined whether SI induces short-term alterations in behavioral and neural markers of WM processing. To this end, fifty Chinese–English bilingual interpreting trainees were randomly assigned to a simultaneous interpreting group (SI; n = 26) or a text-comprehension group (TC; n = 24). Participants completed a validated non-verbal visuospatial WM task before and after the exertion phase, while EEG was recorded. Importantly, the task included a recall-only condition assessing unimodal maintenance and an integration–recall condition assessing multimodal updating and feature binding. During the exertion phase, participants either interpreted a naturalistic Chinese speech into English in real time or listened to the same material for comprehension. Behavioral performance was analyzed using linear mixed-effects models, whereas EEG analyses combined cluster-based permutation statistics for event-related potentials (ERPs) and time–frequency representations (TFRs). Behaviorally, the analyses revealed a significant interaction between group and phase in the integration–recall condition [F(1,46) = 4.84, p = .02]. Specifically, whereas the TC group showed practice-related improvements, the SI group did not. Post-hoc analyses further demonstrated comparable pre-exertion performance between groups (p = .24), but significantly lower post-exertion accuracy in the SI group relative to the TC group (p = .02). In contrast, no significant behavioral effects emerged in the recall-only condition. At the neurophysiological level, ERP analyses revealed exertion-related modulations selectively in the SI group. In the integration–recall condition, a significant centro-parietal negative cluster emerged between 238 and 272 ms (t-sum = −657.75, cluster-level p = .022). Likewise, the recall-only condition showed a significant centro-parietal cluster between 248 and 308 ms (t-sum = −1150.30, cluster-level p = .005), both reflecting larger post-exertion negativities. Notably, no significant ERP clusters emerged in the TC group. Complementing these findings, time–frequency analyses revealed reduced beta-band power (16–30 Hz) in the SI group during the integration–recall condition between 300 and 800 ms over widespread centro-fronto-temporal electrodes (t-sum = −20902, cluster-level p = .008). By contrast, no significant oscillatory effects emerged in the recall-only condition or in the TC group. Importantly, the magnitude of beta-band modulation correlated with L1-to-L2 SI competence (r = .46, p = .04), indicating reduced exertion-related beta desynchronization in participants with higher interpreting competence. Taken together, these findings suggest that SI induces short-term neurophysiological alterations in WM systems, selectively impacting updating and integration mechanisms. More broadly, the results indicate that intense bilingual language control transiently taxes WM-related neural dynamics, as reflected in early ERP modulations and sustained beta-band reductions following interpreting exertion.

Topic Areas: Multilingualism,