Neural bases of language switching in high and early proficient bilinguals
Highlights
► Language switching in early bilinguals is related to the left caudate and pre-SMA. ► Brain areas involved in language switching in early and late bilinguals are different. ► The left caudate is specifically related to the inhibition of the dominant language. ► The pre-SMA/ACC participates in backward switching in early bilinguals.
Introduction
The neural implementation of a bilingual’s two languages has been the focus of an important amount of research (Abutalebi, 2008; Indefrey, 2006 for two extensive reviews). However, neuroimaging data on this issue have often lacked information regarding the age of acquisition, the degree of proficiency, and the degree of exposure to a given language of the tested sample, possibly creating confounds in the interpretation of the results (Abutalebi, Tettamanti, & Perani, 2009). It is the aim of the present study to bring new experimental evidence about the neural substrates of language control in speech production in early and high-proficient bilinguals.
High-proficient bilinguals are able to produce and understand complex sentences in a language different from their own mother tongue. In early bilinguals, this ability develops since early childhood, and can be retained and practiced during adulthood. Furthermore, in some sociolinguistic contexts such as the Spanish communities, where Catalan is spoken, there are often bilingual conversations, in which the two languages are concomitantly used according to the interlocutor. This situation leads to a life-long experience with language switching in individuals with a high proficiency in both languages, an activity that it is at the core of language control. The sample of bilinguals explored in our study corresponds to this sort of speakers, namely, individuals that have been exposed to two languages very early in life, use the two of them on everyday basis, and have a very good command in both of them.
The issue of the neural underpinnings of language control has captured the attention of quite a few studies (Crinion et al., 2006, Price et al., 1999; see Abutalebi, 2008, for a review). When considering these studies, irrespective of whether they involve comprehension or production processes, it appears that the language control network involves frontal and prefrontal areas, left inferior and superior parietal cortices, ACC, and the caudate nucleus. These areas have been shown to support general executive control processes (Dehaene and Changeux, 1991, Desimone and Duncan, 1995, Miller and Cohen, 2001). Beyond this general approach, it is important to asses how this general network may be differentially engaged during comprehension or production processes. This is because the involvement of the speaker in language switching in production and comprehension is very different. In production, the speaker is the one that decides whether or not (and when) to switch languages, and hence the switching performance is mostly pro-actively driven. In contrast, in comprehension the listener is a passive receptor of language switching, and hence the switching performance is mostly re-actively driven. In the present study we focus on the bilingual language control system in speech production.
Perhaps the most used paradigm to explore the cognitive mechanisms behind bilingual language control in speech production is the language switching paradigm (Costa and Santesteban, 2004, Costa et al., 2006, Gollan and Ferreira, 2009, Meuter and Allport, 1999), where participants are asked to name pictures in both of their two languages in an intermixed fashion. The language in which a given stimulus has to be named is prompt by a cue; leading to two types of trials: (a) trials in which a response has to be given in the same language as in the immediately preceding trial (non-switch trials) and (b) trials in which the language of the response is different from the one used in the immediately preceding trial (switch trials). The difference in performance between switch and non-switch trials is commonly referred to as the language switch cost. From the behavioural aspect of this research, an important difference associated with language proficiency arises: while low proficient bilinguals show asymmetrical switching costs (a larger cost when switching into L1 – backward switching – than when switching into L2 – forward switching–), early and high-proficient bilinguals show symmetrical switching costs (the same cost when switching into L1 or L2) (Costa and Santesteban, 2004, Costa et al., 2006). This different pattern of switching costs (symmetrical vs. asymmetrical) has been interpreted as revealing a qualitative shift in the cognitive mechanisms responsible for bilingual language control, linked to an increase in proficiency in the non-dominant language (but see: Christoffels, Firk, & Schiller, 2007).
The qualitatively different performance in the language switching task of low and high-proficient bilinguals prompts the hypothesis that the neural circuits sustaining language control in speech production might also be influenced by second language proficiency. Present data, however, do not allow us to answer this issue. This is because most of the research on the neural basis of language switching either involved L2 learners (or late proficient bilinguals; e.g., Abutalebi et al., 2008, Crinion et al., 2006, Price et al., 1999) or made use of language comprehension tasks. Hence, we lack the necessary set of data informing us about whether language proficiency affects the neural substrates of bilingual language control in speech production. It is the goal of the present study to provide some of this data.
Nevertheless, we do have some information about the neural correlates of language switching in speech production in low proficient bilinguals. In Wang, Xue, Chen, Xue, and Dong (2007) study, language switching (irrespective of the switch direction) yielded activations in the left superior and medial prefrontal cortex, and the right caudate. However, there were instructive differences associated with the switching direction. Activations in the left ACC and medial prefrontal areas were associated with forward switching (from L1 to L2), whereas increases of activity in right parahippocampus, left cerebellum, and thalamus were observed for backward switching (from L2 to L1). In a subsequent study, Wang, Kuhl, Chen, and Dong (2009) were able to confirm the involvement of the left superior and medial prefrontal cortex in language switching. However, they failed to detect any involvement of subcortical structures, rendering the role of the caudate in language control in speech production controversial.
Beyond this controversy, our knowledge of what are the specific processes sustained by each brain area during language control is still limited (see Abutalebi & Green, 2007; Wang et al., 2009, for a discussion of this issue). In this respect, it is useful to bring up the distinction between sustained and transient control processes. According to a recent study by Wang et al. (2009) (see also Christoffels et al., 2007) sustained language control, as indexed by the difference in naming situations in which one language is only used and those in which the two languages are at play, would involve bilateral frontal executive regions. In contrast, transient language control, as indexed by the difference between switch and non-switch trials, would involve a more left lateralized fronto-parietal executive network. In the present study, we are interested in evaluating the brain circuits involved in this transient language control, in early and high-proficient bilinguals. Consequently, we will use a design in which the two languages are intermixed in the same testing session, therefore avoiding any effects of sustained language control.
To recapitulate, our study intends to complement and extend Wang et al.’s results by exploring the language control system put at play during the production of speech by early and very high-proficient bilinguals with wide experience in both languages from birth. We do so by analysing the neural correlates associated with language switching in overt speech production. Given that this type of bilinguals, in comparison to late and low proficient bilinguals, exhibit a different pattern of behavioural switching costs (symmetrical vs. asymmetrical), it is reasonable to expect differences in the brain networks associated with language control in the two populations. In particular, we hypothesize that the large differences observed by Wang et al. (2007) between forward and backward switching, will be different in the case of high and very early proficient bilinguals.
Section snippets
Subjects
A total of 19 right-handed university students at the University Jaume I of Castelló were selected for this study: 7 males (mean age 20.3 years, s.d. 2.3, range 18–23) and 12 females (mean age 20.3, s.d. 2.4, range 17–24). The subjects gave written informed consent to the experiment, according to the ethical protocol of the University. Participation in this experiment was monetarily rewarded. All candidates underwent a preliminary interview to ensure they had no physical or psychological
Behavioural results
Giving that all participants were very proficient in both Catalan and Spanish, naming performance in both languages was very high. Most of the errors were omissions, instances in which the participant did not respond: 2.5% in Catalan and 2.3% in Spanish. We did not include in the commission errors, those naming instances in which borrowings from one language into another were produced, regardless of whether these borrowings were normatively accepted. The mean percentage of commission errors was
Discussion
The aim of the present study was to investigate the brain mechanisms involved in language switching in speech production in early and high-proficient bilinguals. It is important to recall a key property of the bilingual participants included in our study. These bilinguals began to learn both languages from birth and were very fluent in both of them. Furthermore, due to the social requirements of their community they are frequently involved in situations requiring language switching as a
Acknowledgements
We would lie to thank the contribution of an anonymous reviewer to this manuscript. This study was supported by two grants of the Spanish Ministry of Science and Technology given to Cesar Avila [the BrainGlot Project, CSD2007-00012 (funded by the Consolider-Ingenio 2010 programme), and the grant PSI2010-20168 and two grants given to Albert Costa (PSI2008-01191 from the MICINN and SGR 2009-1521 from the Catalan Government). The authors wish to thank Ian FitzPatrick for his comments on the
References (29)
Neural aspects of second language representation and language control
Acta Psychologica
(2008)- et al.
Bilingual language production: The neurocognition of language representation and control
Journal of Neurolinguistics
(2007) - et al.
The bilingual brain: Linguistic and non-linguistic skills
Brain and Language
(2009) - et al.
Bilingual language control: An event–related brain potential study
Brain Research
(2007) - et al.
Lexical access in bilingual speech production: Evidence from language switching in highly-proficient bilinguals and L2 learners
Journal of Memory and Language
(2004) - et al.
fMRI evidence that the neural basis of response inhibition is task-dependent
Brain Research. Cognitive Brain Research
(2003) - et al.
Sustained and transient language control in the bilingual brain
Neuroimage
(2009) - et al.
Neural bases of asymmetric language switching in second-language learners: An ER-fMRI study
Neuroimage
(2007) - et al.
Language control and lexical competition in bilinguals: An event/related fMRI study
Cerebral Cortex
(2008) - et al.
The neural cost of the auditory perception of language switches: An event–related fucntional magnetic resonance imaging study in bilinguals
Journal of Neuroscience
(2007)