The concept of inhibition in bilingual control.

Abstract

To achieve fluent language processing as a bilingual, a dominant theoretical framework assumes that the nontarget language is inhibited. This assumption is based on several empirical effects that are typically explained with inhibitory control. In the current article, we discuss four prominent effects linked to bilingual inhibition in language production (i.e., asymmetrical switch costs, n−2 language repetition costs, reversed language dominance, and the blocked language order effect). We argue that these effects require more empirical examination in order to arrive at a firmer basis for the assumption that inhibition plays a major role during bilingual language control. In particular, the empirical replicability of the phenomena themselves needs to be established more firmly, the underlying theoretical assumptions need further examination, and the alternative explanations of the empirical effects need to be scrutinized. In turn, we conclude that inhibitory control may provide a coherent framework for bilingual language production while outlining the challenges that the inhibition account still needs to face.

Tracking components of bilingual language control in speech production: an fMRI study using functional localizers

“…Explanations that assume the involvement of proactive control have been previously used to account for the L2 after-effect Declerck & Koch, 2022), and our current findings are generally consistent with this framework. However, they also point to an important caveat regarding the level of the targeted representations.…”

“…While these accounts provide plausible explanations of this effect, no univocal evidence in favor or against these interpretations has been provided (Casado et al, 2022; Wodniecka, Szewczyk, et al, 2020; Declerck & Koch, 2022). All available accounts of the L2 after-effect build on a common theoretical assumption that the difficulty in producing words in L1 after using L2 is a consequence of a control mechanism that changes the balance between the relative activation of L1 and L2 representations (Casado et al, 2022; for a discussion see Declerck & Koch, 2022). Importantly, according to these accounts the change in L1-L2 balance results in increased interference between the two languages that persists for some time after the speaker switches back to speaking in L1 (Casado et al, 2022), which, in turn, increases the difficulty of access and selection of (a) lexico-semantic representations (e.g., Casado et al, 2022) and/or (b) phonological representations and articulatory programs of L1 (e.g., Guo et al, 2011).…”

“…representations (Casado et al, 2022; for a discussion see Declerck & Koch, 2022). Importantly, according to these accounts the change in L1-L2 balance results in increased interference between the two languages that persists for some time after the speaker switches back to speaking in L1 (Casado et al, 2022), which, in turn, increases the difficulty of access and selection of (a) lexico-semantic representations (e.g., Casado et al, 2022) and/or (b) phonological representations and articulatory programs of L1 (e.g., Guo et al, 2011).…”

No evidence for a mixing benefit—A registered report of voluntary dialect switching

“…Along the same lines, it is still unclear what the boundary conditions are to observe asymmetrical switch costs (for reviews, see [ 4 , 27 , 58 ]). So, finding symmetrical switch costs might also be due to characteristics of this study other than voluntary language switching.…”

Proactive and reactive language control in bilingual language production revealed by decoding sustained potentials and electroencephalography oscillations

“…Consistent with previous findings (e.g., Costa & Santesteban, 2004;Declerck et al, 2015;Gollan & Ferreira, 2009;Li & Gollan, 2018), this result suggests that unbalanced bilinguals exerted strong inhibition on the dominant language during the mixed language naming condition. It should be noted that there are different interpretations of the reversed language dominance effect: While some researchers argue it is a strong indicator of inhibition over the dominant language (e.g., Bobb & Wodniecka, 2013;Fu et al, 2017), other scholars suggest it could also be explained with alternative mechanisms (see Gade et al, 2021aGade et al, , 2021b for related discussions), such as persistent over-activation (e.g., Declerck et al, 2015;Declerck & Koch, 2023) or lower selection threshold for the weaker L2 (Costa & Santesteban, 2004). We suppose this effect reflects inhibition over the L1, as it converges with our ERP decoding data discussed below.…”