Abstract
When bilinguals frequently switch between their first (L1) and second (L2) languages during speech production, we usually observe two phenomena: (i) language switch cost , where switching to a different language is more difficult than staying in the same one, and (ii) reversed language dominance , where L1 production becomes slower than L2 production. These effects are thought to reflect language control mechanisms, yet the underlying neural bases remain debated. In this study, we addressed this question by using the precision functional magnetic resonance imaging (fMRI) based on functional localization. Forty-one Polish–English bilinguals performed a language switching task (LST), in which they named pictures in L1 or L2 based on color cues. We investigated mechanisms behind two indices of language control commonly observed in the LST. First, we asked whether the domain-general resources supporting language switch cost overlap with nonverbal task switch cost . Second, we asked whether reversed language dominance reflects changes in language activation in the language-specific system, or whether it is related to increased engagement of domain-general control mechanisms. Results indicated that the language switch cost and nonverbal task switch cost share overlapping domain-general neural mechanisms. Similar to the language switch cost , reversed language dominance primarily engages domain-general processes rather than language-specific resources. Highlights fMRI combined with functional localization approach is implemented to examine the neural mechanisms underlying language switch cost and reversed language dominance . Language switch cost relies on neural mechanisms shared with nonverbal switch cost within the Multiple Demand network. Reversed language dominance is primarily supported by the domain-general rather than the language-specific mechanisms. Domain-general neural mechanisms play a pivotal role in bilingual language switching in speech production.
Full text
2,073 characters
· extracted from
oa-doi-fallback
· click to expand
Abstract
When bilinguals frequently switch between their first (L1) and second (L2) languages during speech production, we usually observe two phenomena: (i) language switch cost, where switching to a different language is more difficult than staying in the same one, and (ii) reversed language dominance, where L1 production becomes slower than L2 production. These effects are thought to reflect language control mechanisms, yet the underlying neural bases remain debated.
In this study, we addressed this question by using the precision functional magnetic resonance imaging (fMRI) based on functional localization. Forty-one Polish–English bilinguals performed a language switching task (LST), in which they named pictures in L1 or L2 based on color cues. We investigated mechanisms behind two indices of language control commonly observed in the LST. First, we asked whether the domain-general resources supporting language switch cost overlap with nonverbal task switch cost. Second, we asked whether reversed language dominance reflects changes in language activation in the language-specific system, or whether it is related to increased engagement of domain-general control mechanisms. Results indicated that the language switch cost and nonverbal task switch cost share overlapping domain-general neural mechanisms. Similar to the language switch cost, reversed language dominance primarily engages domain-general processes rather than language-specific resources.
Highlights
fMRI combined with functional localization approach is implemented to examine the neural mechanisms underlying language switch cost and reversed language dominance.
Language switch cost relies on neural mechanisms shared with nonverbal switch cost within the Multiple Demand network.
Reversed language dominance is primarily supported by the domain-general rather than the language-specific mechanisms.
Domain-general neural mechanisms play a pivotal role in bilingual language switching in speech production.
Competing Interest Statement
The authors have declared no competing interest.
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.