Language context flexibly modulates language control mechanisms.

Does a bilingual's language environment, whether L1- or L2-dominant, modulate their use of language control mechanisms in speech production? The language-switching task (LST) typically assesses two indices of language control: asymmetric switch costs, where switching into L1 incurs greater costs than switching into L2, and reversed language dominance, where L1 becomes slower than L2. We asked if these measures are affected by the language context a bilingual is in and whether the relative balance between languages is sustained after mixed language use. Polish-English unbalanced bilinguals performed the LST in L1- and L2-dominant contexts. We found that language context modulated the magnitude of both language control indices. In the L1-dominant context, we replicated effects typical for unbalanced bilinguals: asymmetric switch costs and reversed language dominance. However, in the L2-dominant context, the asymmetry of switch costs was reversed, and the reversed dominance effect decreased. Furthermore, to assess whether these language context modulations of control indices persisted after leaving the switching task, participants named pictures in pure language blocks before (pre-LST) and after (post-LST) LST. Initially, unbalanced bilinguals exhibited a typical L1 dominance (i.e., faster L1 than L2 naming), which reversed during LST and remained reversed post-LST. The modulations of bilingual language control suggest that immediate language context flexibly shapes control mechanisms for effective production, with adjustments persisting beyond the L1- or L2-dominant context. Thus, mechanisms of control are highly adaptive to contextual demands rather than solely determined by language proficiency or other fixed traits of speakers.