Although bilinguals rarely make random errors of language when they speak, research on spoken production provides compelling evidence to suggest that both languages are active when only one language is spoken (e.g., Poulisse, 1999). Moreover, the parallel activation of the two languages appears to characterize the planning of speech for highly proficient bilinguals as well as second language learners. In this paper we first review the evidence for cross-language activity during single word production and then consider the two major alternative models of how the intended language is eventually selected. According to language-specific selection models, both languages may be active but bilinguals develop the ability to selectively attend to candidates in the intended language. The alternative model, that candidates from both languages compete for selection, requires that crosslanguage activity be modulated to allow selection to occur. On the latter view, the selection mechanism may require that candidates in the non-target language be inhibited. We consider the evidence for such an inhibitory mechanism in a series of recent behavioral and neuroimaging studies.

The current study examined the neural correlates associated with local and global inhibitory processes used by bilinguals to resolve interference between competing responses. Two groups of participants completed both blocked and mixed picture naming tasks while undergoing functional magnetic resonance imaging (fMRI). One group first named a set of pictures in L1, and then named the same pictures in L2. The other group first named pictures in L2, and then in L1. After the blocked naming tasks, both groups performed a mixed language naming task (i.e., naming pictures in either language according to a cue). The comparison between the blocked and mixed naming tasks, collapsed across groups, was defined as the local switching effect, while the comparison between blocked naming in each language was defined as the global switching effect. Distinct patterns of neural activation were found for local inhibition as compared to global inhibition in bilingual word production. Specifically, the results suggest that the dorsal anterior cingulate cortex (ACC) and the supplementary motor area (SMA) play important roles in local inhibition, while the dorsal left frontal gyrus and parietal cortex are important for global inhibition.

Behavioral and event-related potential (ERP) measures are reported for a study in which relatively proficient Chinese-English bilinguals named identical pictures in each of their two languages. Production occurred only in Chinese (the first language, L1) or only in English (the second language, L2) in a given block with the order counterbalanced across participants. The repetition of pictures across blocks was expected to produce facilitation in the form of faster responses and more positive ERPs. However, we hypothesized that if both languages are activated when naming one language alone, there might be evidence of inhibition of the stronger L1 to enable naming in the weaker L2. Behavioral data revealed the dominance of Chinese relative to English, with overall faster and more accurate naming performance in L1 than L2. However, reaction times for naming in L1 after naming in L2 showed no repetition advantage and the ERP data showed greater negativity when pictures were named in L1 following L2. This greater negativity for repeated items suggests the presence of inhibition rather than facilitation alone. Critically, the asymmetric negativity associated with the L1 when it followed the L2 endured beyond the immediate switch of language, implying long-lasting inhibition of the L1. In contrast, when L2 naming followed L1, both behavioral and ERP evidence produced a facilitatory pattern, consistent with repetition priming. Taken together, the results support a model of bilingual lexical production in which candidates in both languages compete for selection, with inhibition of the more dominant L1 when planning speech in the less dominant L2. We discuss the implications for modeling the scope and time course of inhibitory processes.

No abstract