Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Bilingual Language Context Modulates Neural Indices of Executive Control in Cantonese–English Bilinguals
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Yuqi Wang1, Haoyun Zhang2, Quentin Zhen Qin1; 1The Hong Kong University of Science and Technology, 2University of Macau
The relationship between bilingual language control and domain-general executive control has often been examined through cross-task conflict adaptation. Namely, the control demands induced by a preceding language task may modulate performance on a subsequent non-linguistic task. Previous work with highly balanced Welsh–English early bilinguals showed that a mixed-language context reduced the Flanker congruency effect, suggesting enhanced executive control after language switching. However, a similar study testing Mandarin–English late bilinguals, whose L2-English proficiency was much lower (scored 3.8 out of 7) than L1-Mandarin, did not find the same effect. The mixed findings may partly reflect differences in bilingual profile (i.e., L2 proficiency). Therefore, we tested a more balanced, higher-L2 proficiency group, that is, Cantonese–English bilinguals in Hong Kong. They are more balanced between their two languages while maintaining a language-distance profile broadly comparable to that of Mandarin–English bilinguals. The participants, with self-rated English proficiency averaged at 5.14 on a 1–7 scale and a mean age of English acquisition of 3.1 years, completed three counterbalanced language-context blocks: Cantonese-single, English-single, and mixed. Each block began with an auditory picture–word matching task (64 trials) designed to establish the language context, followed immediately by a 64-trial Flanker task. In the language-context phase, participants judged whether a word matched a visually presented picture. Single-language conditions contained words from only one language, whereas the mixed-language conditions contained 32 Cantonese and 32 English words (pseudo-randomly intermixed). In the Flanker task, participants identified the direction of a central arrow flanked by congruent or incongruent arrows. Reaction times in the Flanker were analysed. EEG was recorded during the Flanker task, with analyses focusing on P3 (350–700 ms) amplitude at centro-parietal electrodes. While the data collection is going on, the behavioural results from 45 participants showed robust Flanker congruency effects in all language conditions (β = 0.01, SE = 0.07, t = 14.13, p < .001), with slower responses to incongruent than to congruent trials. This congruency effect was numerically the smallest after the mixed condition. The EEG results provided robust evidence for language-context modulation. The P3 amplitude showed significant congruency difference after both Cantonese-single (β = 4.69, SE = 0.66, z = 7.08, p < .001) and English-single conditions (β = 1.60, SE = 0.68, z = 2.34, p = .019), whereas this difference was absent after the mixed condition (β = -0.41, SE = 0.66, z = -0.62, p = .532). The results showed that the mixed condition reduced neural sensitivity to conflict in the Flanker task, compared with single-language conditions. For Cantonese–English bilinguals, the mixed condition pre-engaged executive control mechanisms, and this control state carried over to subsequent Flanker tasks. The current sample of bilinguals had relatively homogeneous L2 proficiency, such that variability in proficiency did not produce detectable differences in cross-task adaptation in the statistical analysis. While more participants with a broader proficiency range will be tested, the findings reconcile previous mixed evidence by suggesting that the language context effect on executive control may be modulated by bilingual language profiles.
Topic Areas: Control, Selection, and Executive Processes, Multilingualism