Chapter 7. When L1 suffers
Bilingual language selection and control is often studied in the language switching paradigm, where participants switch between producing words in one or the other language. Although naming in L1 is usually fastest, in language switching studies L1 naming is sometimes found to be slower than L2 naming (e.g., Christoffels, Firk, & Schiller, 2007; Costa & Santesteban, 2004; Gollan & Ferreira, 2009). This “reversed language effect” may be explained by sustained and global inhibition of the L1 in a mixed language context. We studied the scope and duration of this effect by addressing the consequences of mixed language use on later L1 production. Unbalanced bilinguals named the same pictures in Dutch (L1) and English (L2) before and after a central language switching task. In Experiment 1, after an extensive switching task, a reversed language effect was obtained: L1 naming became slower than L2 naming. In Experiment 2 we replicated this reversed language effect and showed that L1 slowing (a) is not limited to the set of words used in the central switching task and (b) is still present after a 10-minutes break. This global and sustained effect on L1 is a clear signature of language control that may be taken as support for sustained L1 inhibition.