Presentation Type
Poster Presentation
Category
STEM (science, technology, engineering, mathematics)
Abstract/Artist Statement
Bilingual language processing requires the coordinated engagement of large-scale brain networks, yet how these networks dynamically reorganize across native (L1) and second language (L2) use—and how this reorganization depends on language proficiency—remains poorly understood. Traditional fMRI studies often characterize brain connectivity as static, overlooking moment-to-moment fluctuations in network interactions that may be critical for successful second-language processing.
In this study, we investigated dynamic brain network organization in Spanish–Basque bilinguals performing a semantic animacy judgment task in their native (Spanish) and second (Basque) language. Using Independent Component Analysis (ICA), we identified large-scale functional networks engaged during L1 and L2 processing and mapped these components onto the canonical Yeo 7-network parcellation. To capture temporal dynamics, we extracted component time courses and applied dynamic functional connectivity analysis combined with Hidden Markov Models (HMMs), enabling the identification of recurring brain states and quantification of their temporal properties, including mean dwell time, fractional occupancy, and transition rates.
Our results reveal clear proficiency-dependent differences in network dynamics. During L2 processing, advanced bilinguals exhibited longer dwell times and fewer state transitions, indicating more stable and efficient network configurations. In contrast, intermediate bilinguals showed shorter dwell times and more frequent transitions, reflecting greater cognitive effort and reduced processing automaticity. Cross-language network similarity analyses further demonstrated stronger alignment between L1 and L2 network configurations in advanced speakers, suggesting a more integrated and automatized bilingual neural architecture. Intermediate speakers showed more variable and distributed network coupling, particularly involving attention, visual, and control networks.
Together, these findings support the Neurocognitive Adaptation Hypothesis, demonstrating that increasing bilingual proficiency is associated with greater stability, efficiency, and convergence of large-scale brain networks during second-language processing. This work highlights the importance of dynamic network approaches for understanding how the bilingual brain adapts to linguistic demands and advances our understanding of neural plasticity in language learning.
Mentor Name
Lucy Owen
Functional Neuroarchitecture of Brain Activation and Network Organization in Bilinguals During Native and Second Language Processing
UC North Ballroom
Bilingual language processing requires the coordinated engagement of large-scale brain networks, yet how these networks dynamically reorganize across native (L1) and second language (L2) use—and how this reorganization depends on language proficiency—remains poorly understood. Traditional fMRI studies often characterize brain connectivity as static, overlooking moment-to-moment fluctuations in network interactions that may be critical for successful second-language processing.
In this study, we investigated dynamic brain network organization in Spanish–Basque bilinguals performing a semantic animacy judgment task in their native (Spanish) and second (Basque) language. Using Independent Component Analysis (ICA), we identified large-scale functional networks engaged during L1 and L2 processing and mapped these components onto the canonical Yeo 7-network parcellation. To capture temporal dynamics, we extracted component time courses and applied dynamic functional connectivity analysis combined with Hidden Markov Models (HMMs), enabling the identification of recurring brain states and quantification of their temporal properties, including mean dwell time, fractional occupancy, and transition rates.
Our results reveal clear proficiency-dependent differences in network dynamics. During L2 processing, advanced bilinguals exhibited longer dwell times and fewer state transitions, indicating more stable and efficient network configurations. In contrast, intermediate bilinguals showed shorter dwell times and more frequent transitions, reflecting greater cognitive effort and reduced processing automaticity. Cross-language network similarity analyses further demonstrated stronger alignment between L1 and L2 network configurations in advanced speakers, suggesting a more integrated and automatized bilingual neural architecture. Intermediate speakers showed more variable and distributed network coupling, particularly involving attention, visual, and control networks.
Together, these findings support the Neurocognitive Adaptation Hypothesis, demonstrating that increasing bilingual proficiency is associated with greater stability, efficiency, and convergence of large-scale brain networks during second-language processing. This work highlights the importance of dynamic network approaches for understanding how the bilingual brain adapts to linguistic demands and advances our understanding of neural plasticity in language learning.