PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of scirepAboutEditorial BoardFor AuthorsScientific Reports
 
Sci Rep. 2017; 7: 11763.
Published online 2017 September 18. doi:  10.1038/s41598-017-12116-w
PMCID: PMC5603581

Bilingual Cortical Control of Between- and Within-Language Competition

Abstract

The human capacity to master multiple languages is remarkable and leads to structural and functional changes in the brain. Understanding how the brain accommodates multiple languages simultaneously is crucial to developing a complete picture of our species’ linguistic capabilities. To examine the neural mechanisms involved in processing two languages, we looked at cortical activation in Spanish-English bilinguals in response to phonological competition either between two languages or within a language. Participants recognized spoken words in a visual world task while their brains were scanned using functional magnetic resonance imaging (fMRI). Results revealed that between-language competition recruited a larger network of frontal control and basal ganglia regions than within-language competition. Bilinguals also recruited more neural resources to manage between-language competition from the dominant language compared to competition from the less dominant language. Additionally, bilinguals’ activation of the basal ganglia was inversely correlated with their executive function ability, suggesting that bilinguals compensated for lower levels of cognitive control by recruiting a broader neural network to manage more difficult tasks. These results provide evidence for differences in neural responses to linguistic competition between versus within languages, and demonstrate the brain’s remarkable plasticity, where language experience can change neural processing.

Introduction

One of the most groundbreaking and exciting findings in neuroscience in recent decades is the extent to which experience changes the brain. Whether it is due to musical training1, spatial navigation2, or motor skill development3, the brain remains highly plastic throughout life. Of all human experiences, language use has arguably the greatest potential to generate widespread neural changes, given its fundamental role in human function and the broad neural network essential to language processing. One example of a linguistically and cognitively demanding experience that causes substantial structural4,5 and functional4,6 changes in the brain is bilingualism. Bilinguals’ ability to switch between languages and maintain separation between them may appear effortless, but this seamlessness masks a complex interactive cognitive architecture. The aim of the current investigation is to examine the neural mechanisms enabling bilinguals to control their two languages during spoken language comprehension.

The need to maintain two languages provides unique challenges for bilinguals and affects neural processing of language and executive function. For example, behavioral and neuroimaging studies have shown that when bilinguals perform a language comprehension task in one language, the other non-target language also becomes activated79. Because bilinguals’ two languages are consistently coactivated, bilinguals’ brains adapt to manage language access differently than monolinguals. This experience managing linguistic activation appears to also extend to non-linguistic tasks. For example, bilinguals’ use of executive function in non-linguistic attentional control tasks has revealed differences compared to monolinguals in the recruitment of structures such as the prefrontal cortex and the anterior cingulate cortex1012.

Activation of non-target words during language comprehension is not itself unique to bilinguals. As listeners hear speech input unfold over time, similar-sounding words become activated13. These lexical items compete for selection14,15, and must be inhibited before language comprehension can proceed. Both monolingual14,15 and bilingual listeners16,17 experience lexical competition during spoken language comprehension. Compared to monolinguals who face phonological competition only within a single language, bilinguals experience competition both within one language as well as between their two languages. For example, when Russian-English bilinguals are instructed (in one language only) to find an item on a visual display (e.g., “Click on the shovel”), they make eye movements to objects that are phonologically similar both in English (shark) and in Russian (sharik, Russian for “balloon”)17. Bilinguals’ fixations to both between- and within-language competitor objects, even when only one language is being used, suggest that lexical items in both languages are activated simultaneously.

Due to the highly interactive nature of a bilinguals’ two languages, cognitive control mechanisms are required to overcome competition. This suggests that bilinguals’ management of linguistic competition may be associated with their general executive control abilities12. Indeed, bilinguals have been shown to use general executive control to manage switching between their two languages in production1820 and comprehension21. The link between executive control resources and the management of phonological competition within a single language has also been investigated and revealed that bilinguals recruit an efficient network of control regions to overcome within-language competition22. This efficient deployment of neural resources observed in bilinguals provides additional empirical support for the recently-proposed Bilingual Anterior to Posterior and Subcortical Shift model (BAPSS)4. The BAPSS model proposes that, with increased second language experience, bilinguals shift their processing load during executive tasks from frontal control areas, like the dorsolateral prefrontal cortex and anterior cingulate, to posterior perceptual areas and subcortical regions, such as the basal ganglia.

The present research aims to use phonological competition resolution to look at cortical processing of between- and within-language competition. The neural underpinnings of phonological competition within a single language have been shown in both monolingual and bilingual speakers. For example, English monolingual speakers activate frontal and temporal language regions in response to phonological competition23, particularly the left supramarginal gyrus and the left inferior frontal gyrus. For bilinguals, the neural correlates of overcoming competition arising within a single language have been elucidated22, but the cortical regions subserving between-language competition remain unknown. In response to within-language competition, bilinguals’ brains efficiently recruit frontal executive control regions, as they are able to manage competition without exhausting additional cortical resources relative to when competition is not present. Phonological competition that occurs between two languages, however, may present additional challenges for the bilingual, because of the additional need to control activation of the non-target language. In the current paper, we seek to compare how phonological competition that emerges between two languages and competition arising within a single language are managed in bilinguals.

The two main objectives of the present study are: (1) To compare the neural resources required to overcome between-language versus within-language competition; and (2) To examine the relationship between language coactivation and cognitive control. We predict that when confronted with between-language competition, bilingual listeners will recruit a broader network of frontal-control regions than that required by within-language competition. Because of the link between linguistic and non-linguistic cognitive control, we expect that individual differences in inhibitory control skill will be related to degree of neural activation in response to phonological competition.

Results

To examine the neural mechanisms involved in processing two languages, we compared cortical activation in Spanish-English bilinguals in response to between- versus within-language phonological competition. Participants completed a spoken word recognition task using the visual world paradigm while in a functional magnetic resonance imaging (fMRI) scanner, under four experimental conditions: English within-language (English target, English competitor), English between-language (English target, Spanish competitor), Spanish within-language (Spanish target, Spanish competitor), and Spanish between-language (Spanish target, English competitor). Neural responses to phonological competitors were compared to baseline trials where no phonological competitors were present. In addition, participants completed assessments of phonological working memory, inhibitory control, and language proficiency outside the scanner.

Accuracy and Response Time

For both accuracy and RT, the effects of Competition (competitor, unrelated), Language (target in English or in Spanish), and Type (between-language, within-language) and their interactions were analyzed using logistic (for accuracy) or linear (for RT) mixed effect regression (using the lme4 package24 in the R statistical computing environment25), including subject and item as random effects.

Accuracy was high overall (M = 91.57%, SD = 27.79) and there were no significant main effects, but there was a significant Language by Type interaction (Estimate = 19.4, SE = 9.0, z = 2.161, p < 0.05). Follow-up pairwise comparisons did not reveal any significant differences between the four levels of Language and Type (ps > 0.1). To account for a crossover interaction of Language and Type, we ran a new model including fixed effects of Competition, Target-Language, and Competitor-Language. This follow-up model was used to determine whether the language of the competitor, rather than its relation to the target (i.e., within- or between-languages) affected accuracy. We found a significant main effect of Competitor-Language (Estimate = 9.7, SE = 4.5, z = 2.161, p < 0.05), with lower accuracy for the English-Competitor condition, 86.67%, SE = 0.032, compared to the Spanish-Competitor condition 96.35%, SE = 0.032. These results indicate that as a competing language, English, which was the participants’ dominant language, affected task performance more than Spanish did. Outlier RTs (greater than the global mean plus two standard deviations) were replaced with M + 2.5 SDs (2.80% of trials). RTs on correct trials were 1785.18ms (SD = 383.35) overall. No significant main effects or interactions were observed (the same pattern of results was observed in an analysis removing outlier RTs entirely, with no significant main effects or interactions).

Functional Neuroimaging

Comparing between-language versus within-language competition

To examine differences in the cortical resources recruited to manage between-language versus within-language phonological competition across languages, we ran a 2 (Condition: between-language, within-language) by 2 (Language: English, Spanish) within-subject ANOVA on the competitor > unrelated-filler contrasts. There was a main effect of Condition; bilinguals recruited more cortical resources to manage between-language competition, where there was increased activation of the left putamen and caudate, as well as the right middle frontal gyrus and superior frontal gyrus (Fig. 1, Table 1C).

Figure 1
Between-language versus within-language phonological competition. Bilinguals showed increased activation in the left putamen and caudate, as well as the right middle frontal gyrus and superior frontal gyrus during between-language competition compared ...
Table 1
Effects of phonological competition between- and within-language.

Between-language competition

Overall effects of between-language competition were assessed using a one-way ANOVA on the contrast comparing between-language competitors and their matched unrelated trials. During between-language competitor trials compared to the no-competition filler trials (between-language competitor > unrelated contrasts), bilinguals showed increased bilateral activation of the middle frontal gyrus and superior frontal gyrus and increased activation of bilateral caudate and putamen (Fig. 2, Table 1A). Comparisons of unrelated > between-language competitor trials showed no increased activation in any regions examined.

Figure 2
Between-language phonological competition versus matched unrelated trials. Bilinguals showed increased activation in bilateral middle frontal gyrus and superior frontal gyrus, as well as bilateral caudate and putamen, during between-language competitor ...

Within-language competition

Overall effects of within-language competition were assessed with a one-way ANOVA on the contrast comparing within-language competitors and their matched unrelated trials. No significant clusters were identified for either the within-language competitor > unrelated or the unrelated > within-language competitor contrasts in any brain regions examined (Table 1B).

Language-specific effects during between-language competition

Planned comparisons on the effect of between-language competition in each language were run using one-way ANOVAs comparing between-language competitors to matched unrelated trials, separately for each language. When auditory cues were received in Spanish, between-language competitor trials resulted in greater activation of bilateral inferior frontal gyrus, middle frontal gyrus, superior frontal gyrus, as well as the caudate and putamen bilaterally, compared to the no-competitor filler trials (Fig. 3, Table 2A). When participants heard the auditory target cue in English, no significant clusters were identified (Table 2B).

Figure 3
Spanish-English between-language phonological competition versus matched unrelated trials. Compared to trials during which there were no competitors, when receiving language input in Spanish and facing competition from English, bilinguals showed increased ...
Table 2
Language-specific effects of between-language phonological competition.

To further analyze language-specific effects of between-language competition, we ran an additional ANOVA comparing the competitor > unrelated-filler contrasts in each language (Spanish-English between-language competition versus English-Spanish between-language competition). When listening to auditory instructions in Spanish and facing between-language competition from English, bilinguals showed increased activation of the right middle frontal gyrus, superior frontal gyrus, and inferior frontal gyrus, compared to when receiving auditory input in English and facing between-language competition from their native language, Spanish (Fig. 4, Table 2C). The English-Spanish between-language competition > Spanish-English between-language competition contrasts did not yield any significant clusters.

Figure 4
Language-specific activation during phonological competition. Bilinguals showed increased activation of the right middle frontal gyrus, superior frontal gyrus, and inferior frontal gyrus during trials where the language input was in Spanish and the between-language ...

Correlations with proficiency and inhibitory control

To examine the relationship between linguistic and non-linguistic inhibitory control, we correlated bilinguals’ Simon inhibition scores and their overall RTs on the Simon task with individual effect sizes in task-activated ROIs for each of the four competitor contrasts (English within-language, Spanish within-language, English-Spanish between-language, and Spanish-English between-language). We also examined the relationship between non-dominant language proficiency and phonological competition by correlating bilinguals’ self-reported proficiency with effect sizes in task-activated ROIs.

Four ROIs were identified for overall between-language competition (Fig. 2), including right middle frontal gyrus (MFG)/superior frontal gyrus (SFG), left MFG/SFG, right putamen, and left putamen. We found that bilinguals with better (smaller) Simon inhibition scores had decreased activation in the right putamen in response to receiving input in Spanish and facing competition from English, R 2 = 0.41, p < 0.05. We also found a marginal correlation between Simon inhibition and activation in the right MFG/SFG, R 2 = 0.21, p < 0.1, where better Simon inhibition was associated with decreased activation of right MFG/SFG in response to receiving input in English and facing competition from Spanish. Further, we found that bilinguals with faster overall Simon RTs displayed decreased left MFG/SFG activation when receiving input in English and facing competition from Spanish, R 2 = 0.62, p < 0.05. There were no correlations between Simon task performance and activation in the left caudate/putamen, and no correlations between proficiency and neural activation.

Two ROIs were identified from the between-language versus within-language competition effect (Fig. 1), including left putamen/caudate and right MFG/SFG. We found a marginal correlation in the right MFG/SFG, R 2 = 0.28, p < 0.1, where better Simon inhibition was associated with decreased activation in response to English input and Spanish competition. No correlations were found with the left putamen/caudate.

Four ROIs were identified from the Spanish-English between-language competitor versus unrelated analysis (Fig. 3), including right MFG/SFG, left MFG/SFG, right putamen/caudate, and left putamen/caudate. When receiving language input in Spanish and facing English competitors, bilinguals with better (smaller) Simon inhibition scores had decreased activation in the right putamen in response to receiving input in Spanish and facing competition from English, R 2 = 0.24, p < 0.1. There were no correlations with Simon inhibition or overall RTs detected in right or left MFG/SFG or in the left putamen/caudate. We found correlations between subcortical activation and proficiency, where greater non-dominant language proficiency was associated with increased activation when receiving input in Spanish and facing competition from English in both left putamen, R 2 = 0.50, p < 0.01, and right putamen, R 2 = 0.25, p < 0.1. One ROI was identified from the Spanish-English between-language competition versus English-Spanish between-language competition analysis (Fig. 4) in right MFG/SFG, but no correlations were found between cortical activation and Simon inhibition or Simon task RTs.

Discussion

Bilinguals’ ability to seamlessly switch between two distinct communication systems masks the considerable control exerted at the neural level. Spoken language comprehension is especially taxing, because the bilingual listener does not control the language of the input, and spoken words unfold gradually over time. To cope with this uncertainty in the input, bilinguals start by activating words in both languages that partially overlap with the auditory signal before selecting the best match. The current study examined the neural networks involved in processing two languages, offering the first comparison of how bilinguals control phonological competition that arises between both of their languages or within one language.

We found that the size and type of the neural network that bilinguals recruited to resolve phonological competition differed depending on the source of competition. When competition occurred between two languages, bilinguals recruited additional frontal control and subcortical regions, specifically the right middle frontal gyrus, superior frontal gyrus, caudate, and putamen, compared to competition that occurred within a single language. This difference in recruitment of brain regions suggests that between-language competition may require additional effort to suppress activation of the non-target language. Increased putamen and caudate activity in the between-language condition relative to the within-language condition is also in line with research suggesting that these areas are involved in cognitive and motor control2629. The neural patterns observed in the present experiment provide additional empirical support for the BAPSS model4 and are consistent with the model’s proposal that between-language competition is initially managed by an extensive network of frontal regions, but later relies less on the frontal network and more on subcortical structures, such as the putamen and caudate. Even within our group of highly proficient bilinguals, the activation of subcortical structures was found to be related to small differences in proficiency. Specifically, as proficiency in the non-dominant language increased, so too did activation in the putamen in response to competition that occurred between languages.

We also found that the neural mechanisms recruited by bilinguals to resolve phonological competition differed depending on the language of competition. Specifically, relative proficiency in the two languages made a greater difference than age of acquisition in determining neural activation. The language with higher proficiency, English, showed more extensive activity than participants’ native language, Spanish. This is consistent with previous literature on bilingual language processing, showing that proficiency is more important than age of acquisition in determining cortical activation30,31. Although counterintuitive, earlier age of acquisition does not always lead to higher proficiency32, as is the case for the participants in the current study, who were Spanish native and English dominant, due to growing up in a country where English was the majority language, including in all academic settings. During between-language competition, when bilinguals received auditory input in Spanish and faced competition from English (the more dominant language), additional frontal control regions were recruited, including the right middle frontal gyrus and superior frontal gyrus, compared to when they received input in English and faced competition from Spanish (the less dominant language). This neural pattern suggests that additional cognitive resources are required to suppress competition from the more dominant language than competition from the less dominant language. These results are in line with cognitive models such as the Revised Hierarchical Model33, which propose that bilinguals access words differently as their second language proficiency increases; specifically, access to the second language may become more automatic with increasing proficiency, rather than controlled. Just as bilinguals with higher second language proficiencies have been shown to have decreased prefrontal activation relative to low-proficiency bilinguals34, bilinguals in the current study also showed less activation of frontal control regions when processing target words in English, which, although acquired second, was their dominant and more automatic and proficient language. Thus, proficiency in a language may be more important than age of acquisition for predicting the degree of competition experienced across languages.

Notably, these neural differences in response to phonological competition were observed despite minimal behavioral effects (i.e., accuracy and RT). In visual world tasks, phonological competition does not always have an effect on accuracy or RT, and is often only detected with more sensitive measures like eye-tracking or mouse-tracking35,36. We did find an interaction between language and competition type in accuracy, where the two conditions that included English competitors (i.e., English within-language and Spanish between-language) were responded to less accurately than the conditions that included Spanish competitors (i.e., Spanish within-language and English between-language). These findings are consistent with the neural results, which indicated that the more proficient and more dominant language exerted a stronger competition effect than the less proficient and less dominant language.

Looking at the functional neuroimaging data, we did not find significant differences in neural activation for within-language competition relative to baseline. There may be two factors contributing to this lack of an effect. First, bilinguals may have more experience resolving within-language competition, as words tend to have more within-language than between-language competitors due to phonological differences across languages. Second, because within-language competition does not increase activation of the non-target language, it may require fewer resources to resolve. There were also no significant differences in brain activation when participants heard English cues and faced competition from Spanish, compared to baseline. This finding could potentially be explained by the source of linguistic competition. Because English is the participants’ more dominant language, facing competition from the less dominant language, Spanish, may be cognitively less taxing and does not require additional neural resources to resolve3739.

Additionally, correlations between bilinguals’ activation of the basal ganglia and Simon inhibition scores indicate that bilinguals with lower cognitive control may compensate by increasing basal ganglia recruitment during more difficult tasks (i.e., between-language competition). This suggests that bilinguals are able to flexibly deploy the necessary neural resources to meet the needs of more complex tasks40,41.

In conclusion, bilinguals differ in how they respond to spoken-word competition, depending on whether the source is between- or within-language, or from the more proficient or less proficient language. Specifically, between-language competition recruits a larger network of frontal control and basal ganglia regions than within-language competition, and competition from the dominant language recruits more neural resources than competition from the less dominant language. Additionally, during more taxing language tasks, bilinguals compensate for lower inhibitory control ability by increasing their activation of the basal ganglia. These findings demonstrate the considerable neural plasticity that enables bilinguals to process speech in spite of linguistic competition from multiple sources.

Methods

Participants

Sixteen Spanish-English bilinguals participated. Data from the English Within-Language competition session for 8 participants were previously reported in Marian et al.22. Participants’ age ranged from 18–32 years, all reported normal or corrected-normal vision and no history of neurological or psychiatric illness, and were right-handed. Bilingual language status was confirmed using the Language Experience and Proficiency Questionnaire (LEAP-Q)42. Bilinguals were exposed to Spanish from birth and to English by the age of 8 (mean age of English acquisition = 5.00 years, SD = 1.69 years); all reported English and Spanish proficiencies of at least a 7 on a 0 (none) to 10 (perfect) scale (mean Spanish proficiency = 8.49, SD = 0.92; mean English proficiency = 9.62, SD = 0.52; t(14) = −4.26, p < 0.05). Participants were English-dominant, were more proficient in English, and were living and attending a university in the United States where English was the official language.

Experimental protocols were approved by the institutional review boards at Northwestern University and University of Houston, concordant to the relevant guidelines. Informed consent was obtained from all participants.

Materials

Four groups of twenty competitor sets were constructed: English within-language (English target, English competitor), English between-language (English target, Spanish competitor), Spanish within-language (Spanish target, Spanish competitor), and Spanish between-language (Spanish target, English competitor). Each English within-language set contained a target word (e.g., candy), an English phonological onset competitor (e.g., candle), and two filler items that did not share overlap with any other items in the set (stimuli from the English within-language set were identical to those used by Marian et al.22). The English between-language stimuli contained an English target (e.g., glue), a Spanish phonological competitor (e.g., globo [balloon]), and two phonologically unrelated fillers. Spanish stimuli were constructed similarly: the Spanish within-language stimuli contained Spanish targets and competitors (e.g., calabaza [pumpkin] – calcetín [sock]), while the Spanish between-language stimuli contained a Spanish target and English competitor (e.g., muñeca [doll] – moon).

In addition to the 80 competitor trials (20 from each condition), 80 unrelated trials were created as a baseline. Unrelated trials were formed by modifying a competitor trial, replacing the competitor with an item that did not overlap with the target. Competitor items were re-used as unrelated items for a different target to control for visual familiarity. In total, the materials included 80 unique Targets, 80 unique Competitors, and 160 unique Fillers which were never repeated across languages or conditions (i.e., within vs. between-language trials). Each picture was viewed two times, once in a competitor trial and once in a matched unrelated trial. See Supplementary Information for a full list of stimuli.

Across all conditions, targets and competitors shared an average of 2.49 (SD = 0.66) phonemes at onset (English within-language: M = 2.40, SD = 0.68; English between-language: 2.55, SD = 0.69; Spanish within-language: M = 2.45, SD = 0.69; Spanish between-language: M = 2.55, SD = 0.60; F(3,76) = 0.25, n.s.). All critical stimuli (targets, competitors, and fillers) were matched on word frequency, orthographic and phonological neighborhood size (CLEARPOND 43), and concreteness, familiarity, and imageability (English: MRC Psycholinguistic Database 44; Spanish: BuscaPalabras 45) across both English and Spanish (all ps > 0.05).

Black and white line drawings were obtained for each object from the International Picture Naming Project (IPNP) database46 based on high naming consistency in both English and Spanish. Items that were unavailable from IPNP were selected from Google Images and independently normed by 20 Spanish-English bilinguals on Amazon Mechanical Turk (https://www.mturk.com). Naming reliability reached 91.48% (SD = 9.88) in English and 84.98% (SD = 15.89) in Spanish.

Following the structure of Marian et al.22, images were placed in the outer four corners of the screen at a visual angle of 13–15°. Target locations were counterbalanced across trials; targets occupied the same quadrant across competitor and matched unrelated trials. Competitors were always located in one of the quadrants adjacent to the target.

The 160 trials were divided into two blocks: an English run (20 English within-language, 20 English between-language, and 40 English unrelated trials), and a Spanish run (20 Spanish within-language, 20 Spanish between-language, and 40 Spanish unrelated trials). Within each block, trials were presented in a pseudo-randomized order (repeated images were separated by at least three trials, and the four quadrants contained an equal number of targets) that was fixed between participants. To control for familiarity effects, half of the participants received the trials in reverse order. Blocks were presented in a counterbalanced order between participants, with half of the participants receiving the English block first and half of the participants receiving the Spanish block first.

Procedure

Testing occurred across two sessions: the first for cognitive and behavioral assessments and the second for administration of the fMRI task. During the behavioral session, tests of phonological working memory (Comprehensive Test of Phonological Processing 47), inhibitory control (Simon Task 48), and language proficiency in both English (Woodcock Language Proficiency Battery-Revised 49) and Spanish (Woodcock-Munoz Language Survey-Revised 50) were administered. See Table 3 for participants’ performance summary.

Table 3
Cognitive and Linguistic Participant Demographics.

During the fMRI session, participants were familiarized with the fMRI scanner and all procedures. Participants were given sound-dampening headphones to reduce scanner noise, a squeeze ball to signal the fMRI technician, and a four-button button box. A display of four images was projected onto a mirrored screen, and participants heard auditory instructions to locate one of the four images.

Timing and trial structure was identical to that used by Marian et al.22. A four-item visual search display was shown for 500 ms before participants heard an auditory target word (spoken by a male Spanish-English bilingual) at 48 Khz, amplitude-normalized. Following the auditory token, the search display remained on the screen for 2500 ms; participants used the button box to indicate target location. Each visual quadrant corresponded to a single button (the bottom left button corresponded to the bottom left quadrant, etc.). Stimuli were presented using E-Prime 2.0 software (Psychology Software Tools, Pittsburgh, PA) in an event-related design with an inter-stimulus interval between 4.5–11.7 seconds.

Following the search task, participants named all competitor pictures in English and Spanish while seated at a computer outside the scanner. Trials in which participants failed to provide a name, or provided an incorrect name were removed from all analyses (6.72% of English trials, 13.6% of Spanish trials).

Neuroimaging Parameters

Functional neuroimaging data were collected using a 3.0 Tesla head-only Siemens Magnetom Allegra magnetic imager located in Baylor College of Medicine’s Human Neuroimaging Laboratory. Anatomical images were acquired using a high-resolution T1-weighted MPRAGE sequence (voxel size = 1.0 × 1.0 × 1.0 mm, TR = 1200 ms, TE = 2.93 ms, reconstructed into 192 slices). Functional images were acquired in 34 axial slices parallel to the AC-PC line with an interleaved descending gradient recalled echo-planar (EPI) imaging sequence (voxel size = 3.4 × 3.4 × 4.0 m, TR = 2700 ms, and TE = 28 ms).

Data analysis

Accuracy and response time

Response time (RT) was measured from the onset of the search display to button response. RTs were analyzed for correct trials only; 2.80% of trials were identified as outliers (longer than the global mean plus 2.5 standard deviations) and were replaced with the threshold value (i.e., M + 2.5 SD). Accuracy and RT were compared across trial types using logistic (for accuracy) or linear (for RT) mixed effect (LME) regression, using the lme4 package24 in the R statistical computing environment25. Models included subject and item random effects, and fixed effects of competition (competitor, unrelated), type (between-language, within-language), and presentation language (English, Spanish). Parameter-specific p values were estimated by using a normal approximation, treating the t value from the model as a z value51. Follow-up pairwise comparisons were performed using Welch t-tests and the Satterthwaite approximation for degrees of freedom, with the Tukey correction for multiple comparisons.

Functional neuroimaging

Functional images were analyzed using SPM8 (Wellcome Trust Centre for Neuroimaging, London, UK). Images were realigned for motion correction, resliced, and slice time corrected. Functional images were coregistered to align with the structural image, segmented, and normalized to a standard Montreal Neurological Institute (MNI) template. Data were spatially smoothed using an 8 mm full-width half maximum (FWHM) Gaussian kernal.

In first-level processing, stimulus onsets locked to the auditory stimulus for the four competitor conditions (within-English competitor, between-English competitor, within-Spanish competitor, between-Spanish competitor) were contrasted against their matched unrelated trial onsets (within-English unrelated, between-English unrelated, within-Spanish unrelated, between-Spanish unrelated) in each participant using a General Linear Model (GLM). Motion estimates from preprocessing were entered as covariates of no interest at the first-level to further control for motion artifacts52. A 2 (condition: between-language, within-language) × 2 (language: English, Spanish) within-subject ANOVA assessed main effects of condition and language on phonological competition. Secondary analyses in each language were performed using one-way within-subject ANOVAs. Based on cortical activation during linguistic competition18,53, analyses were restricted to an ROI including bilateral inferior frontal gyrus, middle frontal gyrus, superior frontal gyrus, inferior parietal lobule, superior parietal lobule, anterior cingulate, and basal ganglia using anatomical definitions in the AAL template54. Monte Carlo simulations with AFNI’s ALPHASIM program were performed to correct for multiple comparisons. All comparisons used a voxel-level threshold of p < 0.025 and a minimum cluster size of 509 contiguous voxels, for a cluster-level significance of p < 0.05.

The relationship between proficiency, inhibitory control, and cortical activation in response to phonological competition was examined using individual effect sizes in task-identified regions of interest (ROIs) following a leave-one-subject-out (LOSO) approach55. Sixteen separate LOSO GLMs were performed, each with n = 15. Task-activated ROIs were identified in each model using the same procedure as the full analysis (ROIs identified in less than a third of LOSO GLMs were not analyzed further). For each participant, mean beta weights for within-English, between-English, within-Spanish, and between-Spanish competition effects were calculated in each ROI from the LOSO GLM that excluded that participant, preserving independence of ROI selection and measured task activation. Mean beta weights were correlated with participants’ language proficiency, Simon inhibition scores (RT on incongruent trials minus RT on neutral trials) and overall Simon task RTs. Simon inhibition scores were used instead of the classic Simon effect (i.e., incongruent minus congruent), as the former provides a more targeted measure of interference suppression56.

Data availability

The datasets analyzed are available from the corresponding author on reasonable request.

Electronic supplementary material

Dataset 1(12K, xlsx)

Dataset 2(107K, xlsx)

Dataset 3(14K, xlsx)

Acknowledgements

The authors would like to thank the members of the Northwestern University Bilingualism and Psycholinguistics Research Group for comments on this work. This research was supported by grant NICHD R01 HD059858 to V.M. and grant NIH/NICHD 1R21 HD059103 to A.E.H.

Author Contributions

Author Contributions

V.M., J.B., K.B., and A.E.H. designed the protocol; V.M. and J.B. designed the stimuli; K.B. and A.E.H. collected the data; J.B. and K.B. analyzed the data; S.R. drafted, and revised the manuscript; J.B. prepared the figures and tables; V.M., S.R., and J.B. prepared the manuscript; A.E.H. and K.B. provided comments on the final document; V.M. coordinated the project and writing.

Notes

Competing Interests

The authors declare that they have no competing interests.

Footnotes

Electronic supplementary material

Supplementary information accompanies this paper at 10.1038/s41598-017-12116-w.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Wan CY, Schlaug G. Music making as a tool for promoting brain plasticity across the life span. Neurosci. 2010;16:566–577. [PMC free article] [PubMed]
2. Maguire EA, et al. Navigation-related structural change in the hippocampi of taxi drivers. Proc. Natl. Acad. Sci. 2000;97:4398–4403. doi: 10.1073/pnas.070039597. [PubMed] [Cross Ref]
3. Ungerleider LG, Doyon J, Karni A. Imaging brain plasticity during motor skill learning. Neurobiol. Learn. Mem. 2002;78:553–564. doi: 10.1006/nlme.2002.4091. [PubMed] [Cross Ref]
4. Grundy, J. G., Anderson, J. A. E. & Bialystok, E. Neural correlates of cognitive processing in monolinguals and bilinguals. Ann. N. Y. Acad. Sci., doi:10.1111/nyas.13333 (2017). [PubMed]
5. Mechelli A, et al. Structural plasticity in the bilingual brain. Nature. 2004;431:3017. doi: 10.1038/431757a. [PubMed] [Cross Ref]
6. Kovelman I, Baker SA, Petitto L-A. Bilingual and monolingual brains compared: a functional magnetic resonance imaging investigation of syntactic processing and a possible ‘neural signature’ of bilingualism. J. Cogn. Neurosci. 2008;20:153–169. doi: 10.1162/jocn.2008.20011. [PMC free article] [PubMed] [Cross Ref]
7. Marian V, Spivey MJ, Hirsch J. Shared and separate systems in bilingual language processing: Converging evidence from eyetracking and brain imaging. Brain Lang. 2003;86:70–82. doi: 10.1016/S0093-934X(02)00535-7. [PubMed] [Cross Ref]
8. Martin CD, Dering B, Thomas EM, Thierry G. Brain potentials reveal semantic priming in both the ‘active’ and the ‘non-attended’ language of early bilinguals. Neuroimage. 2009;47:326–333. doi: 10.1016/j.neuroimage.2009.04.025. [PubMed] [Cross Ref]
9. Thierry G, Wu YJ. Brain potentials reveal unconscious translation during foreign-language comprehension. Proc. Natl. Acad. Sci. 2007;104:12530–12535. doi: 10.1073/pnas.0609927104. [PubMed] [Cross Ref]
10. Abutalebi J, et al. Language proficiency modulates the engagement of cognitive control areas in multilinguals. Cortex. 2013;49:905–911. doi: 10.1016/j.cortex.2012.08.018. [PubMed] [Cross Ref]
11. Bialystok E, et al. Effect of bilingualism on cognitive control in the Simon task: Evidence from MEG. Neuroimage. 2005;24:40–49. doi: 10.1016/j.neuroimage.2004.09.044. [PubMed] [Cross Ref]
12. Coderre EL, Smith JF, van Heuven WJB, Horwitz B. The functional overlap of executive control and language processing in bilinguals. Biling. Lang. Cogn. 2015;19:471–488. doi: 10.1017/S1366728915000188. [PMC free article] [PubMed] [Cross Ref]
13. Marslen-Wilson WD. Functional parallelism in spoken word-recognition. Cognition. 1987;25:71–102. doi: 10.1016/0010-0277(87)90005-9. [PubMed] [Cross Ref]
14. Allopenna P, Magnuson JS, Tanenhaus MK. Tracking the time course of spoken word recognition using eye movements: Evidence for continuous mapping models. J. Mem. Lang. 1998;38:419–439. doi: 10.1006/jmla.1997.2558. [Cross Ref]
15. Tanenhaus MK, Spivey-Knowlton MJ, Eberhard KM, Sedivy JC. Integration of visual and linguistic information in spoken language comprehension. Science (80-). 1995;268:1632–1634. doi: 10.1126/science.7777863. [PubMed] [Cross Ref]
16. Marian V, Spivey MJ. Bilingual and monolingual processing of competing lexical items. Appl. Psycholinguist. 2003;24:173–193. doi: 10.1017/S0142716403000092. [Cross Ref]
17. Marian V, Spivey MJ. Competing activation in bilingual language processing: Within- and between-language competition. Biling. Lang. Cogn. 2003;6:97–115. doi: 10.1017/S1366728903001068. [Cross Ref]
18. Abutalebi J, et al. Language control and lexical competition in bilinguals: An event-related FMRI study. Cereb. Cortex. 2008;18:1496–1505. doi: 10.1093/cercor/bhm182. [PubMed] [Cross Ref]
19. Hernandez AE, Dapretto M, Mazziotta JC, Bookheimer S. Language switching and language representation in Spanish-English bilinguals: An fMRI study. Neuroimage. 2001;14:510–520. doi: 10.1006/nimg.2001.0810. [PubMed] [Cross Ref]
20. Hernandez A, Martinez A, Kohnert K. In search of the language switch: An fMRI study of picture naming in Spanish-English bilinugals. Brain Lang. 2000;73:421–431. doi: 10.1006/brln.1999.2278. [PubMed] [Cross Ref]
21. Abutalebi J, et al. The neural cost of the auditory perception of language switches: an event-related functional magnetic resonance imaging study in bilinguals. J. Neurosci. 2007;27:13762–13769. doi: 10.1523/JNEUROSCI.3294-07.2007. [PubMed] [Cross Ref]
22. Marian V, Chabal S, Bartolotti J, Bradley K, Hernandez AE. Differential recruitment of executive control regions during phonological competition in monolinguals and bilinguals. Brain Lang. 2014;139:108–117. doi: 10.1016/j.bandl.2014.10.005. [PMC free article] [PubMed] [Cross Ref]
23. Righi G, Blumstein SE, Mertus J, Worden MS. Neural systems underlying lexical competition: An eye tracking and fMRI study. J. Cogn. Neurosci. 2009;22:213–224. doi: 10.1162/jocn.2009.21200. [PMC free article] [PubMed] [Cross Ref]
24. Bates DM, Machler M, Bolker BM, Walker SC. Fitting linear mixed-effects models using. J. Stat. Softw. 2014;67:lme4.
25. R Core Team. R: A Language and Environment for Statistical Computing. At https://www.r-project.org (2016).
26. Aron AR, et al. Converging evidence for a fronto-basal-ganglia network for inhibitory control of action and cognition. J. Neurosci. 2007;27:11860–11864. doi: 10.1523/JNEUROSCI.3644-07.2007. [PubMed] [Cross Ref]
27. Graybiel AM. The basal ganglia and cognitive pattern generators. Schizophr. Bull. 1997;23:459–469. doi: 10.1093/schbul/23.3.459. [PubMed] [Cross Ref]
28. Middleton FA, Strick PL. Basal ganglia and cerebellar loops: Motor and cognitive circuits. Brain Res. Rev. 2000;31:236–250. doi: 10.1016/S0165-0173(99)00040-5. [PubMed] [Cross Ref]
29. Stocco A, Yamasaki B, Prat CS. Bilingual brain training: A neurobiological framework of how bilingual experience improves executive function. Int. J. Biling. 2014;18:67–92. doi: 10.1177/1367006912456617. [Cross Ref]
30. Elston-Güttler KE, Paulmann S, Kotz SA. Who’s in control? Proficiency and L1 influence on L2 processing. J. Cogn. Neurosci. 2005;17:1593–1610. doi: 10.1162/089892905774597245. [PubMed] [Cross Ref]
31. Perani D, et al. The bilingual brain: Proficiency and age of acquisition of the second language. Brain. 1998;121:1841–1852. doi: 10.1093/brain/121.10.1841. [PubMed] [Cross Ref]
32. Nichols ES, Joanisse MF. Functional activity and white matter microstructure reveal the independent effects of age of acquisition and proficiency on second-language learning. Neuroimage. 2016;143:15–25. doi: 10.1016/j.neuroimage.2016.08.053. [PubMed] [Cross Ref]
33. Kroll JF, Stewart E. Category interference in translation and picture naming: Evidence for asymmetric connections between bilingual memory representations. J. Mem. Lang. 1994;33:149–174. doi: 10.1006/jmla.1994.1008. [Cross Ref]
34. Abutalebi J, Green DW. Bilingual language production: The neurocognition of language representation and control. J. Neurolinguistics. 2007;20:242–275. doi: 10.1016/j.jneuroling.2006.10.003. [Cross Ref]
35. Bartolotti J. & Marian, V. Language learning and control in monolinguals and bilinguals. Cogn. Sci. 2012;36:1129–1147. doi: 10.1111/j.1551-6709.2012.01243.x. [PMC free article] [PubMed] [Cross Ref]
36. Blumenfeld HK. & Marian, V. Bilingualism influences inhibitory control in auditory comprehension. Cognition. 2011;118:245–257. doi: 10.1016/j.cognition.2010.10.012. [PMC free article] [PubMed] [Cross Ref]
37. Jackson GM, Swainson R, Cunnington R, Jackson SR. ERP correlates of executive control during repeated language switching. Biling. Lang. Cogn. 2001;4:169–178. doi: 10.1017/S1366728901000268. [Cross Ref]
38. Meuter R, Allport A. Bilingual language switching in naming: Asymmetrical costs of language selection. J. Mem. Lang. 1999;40:25–40. doi: 10.1006/jmla.1998.2602. [Cross Ref]
39. Philipp AM, Gade M, Koch I. Inhibitory processes in language switching: Evidence from switching language-defined response sets. Eur. J. Cogn. Psychol. 2007;19:395–416. doi: 10.1080/09541440600758812. [Cross Ref]
40. Festman J, Munte TF. Cognitive control in Russian-German bilinguals. Front. Psychol. 2012;3:115. doi: 10.3389/fpsyg.2012.00115. [PMC free article] [PubMed] [Cross Ref]
41. Festman J, Rodriguez-Fornells A, Münte TF. Individual differences in control of language interference in late bilinguals are mainly related to general executive abilities. Behav. Brain Funct. 2010;6:5. doi: 10.1186/1744-9081-6-5. [PMC free article] [PubMed] [Cross Ref]
42. Marian V, Blumenfeld HK, Kaushanskaya M. The language experience and proficiency questionnaire (LEAP-Q): Assessing language profiles in bilinguals and multilinguals. J. Speech, Lang. Hear. Res. 2007;50:940–967. doi: 10.1044/1092-4388(2007/067). [PubMed] [Cross Ref]
43. Marian V, Bartolotti J, Chabal S, Shook A. CLEARPOND: Cross-Linguistic Easy-Access Resource for Phonological and Orthographic Neighborhood Densities. PLoS One. 2012;7:e43230. doi: 10.1371/journal.pone.0043230. [PMC free article] [PubMed] [Cross Ref]
44. Coltheart M. The MRC psycholinguistic database. Quartely J. Exp. Psychol. 1981;33:497–505. doi: 10.1080/14640748108400805. [Cross Ref]
45. Davis CJ, Perea M. BuscaPalabras: a program for deriving orthographic and phonological neighborhood statistics and other psycholinguistic indices in Spanish. Behav. Res. Methods. 2005;37:665–71. doi: 10.3758/BF03192738. [PubMed] [Cross Ref]
46. Bates, E. et al. Introducing the CRL international picture naming project (CRL-IPNP). Cent. Res. Lang. Newsl. 12 (2000).
47. Wagner, R. K., Torgesen, J. K. & Rashotte, C. A. The comprehensive test of phonological processing. (1999).
48. Simon JR, Rudell AP. Auditory S-R compatibility: The effect of an irrelevant cue on information processing. J. Appl. Psychol. 1967;51:300–304. doi: 10.1037/h0020586. [PubMed] [Cross Ref]
49. Woodcock, R. Woodcock Language Proficiency Battery-Revised (WLPB-R). (Riverside, 1995).
50. Woodcock, R., Muñoz-Sandoval, A., Ruef, M. & Alvarado, C. Woodcock-Muñoz language survey-revised. (Riverside, 2005).
51. Barr DJ, Levy R, Scheepers C, Tily HJ. Random effects structure for confirmatory hypothesis testing: Keep it maximal. J. Mem. Lang. 2013;68:255–278. doi: 10.1016/j.jml.2012.11.001. [PMC free article] [PubMed] [Cross Ref]
52. Johnstone T, et al. Motion correction and the use of motion covariates in multiple-subject fMRI analysis. Hum. Brain Mapp. 2006;27:779–88. doi: 10.1002/hbm.20219. [PubMed] [Cross Ref]
53. Milham MP, et al. The relative involvement of anterior cingulate and prefrontal cortex in attentional control depends on nature of conflict. Brain Res. Cogn. Brain Res. 2001;12:467–73. doi: 10.1016/S0926-6410(01)00076-3. [PubMed] [Cross Ref]
54. Tzourio-Mazoyer N, et al. Automated Anatomical Labeling of Activations in SPM Using a Macroscopic Anatomical Parcellation of the MNI MRI Single-Subject Brain. Neuroimage. 2002;15:273–289. doi: 10.1006/nimg.2001.0978. [PubMed] [Cross Ref]
55. Esterman M, Tamber-Rosenau BJ, Chiu YC, Yantis S. Avoiding non-independence in fMRI data analysis: Leave one subject out. Neuroimage. 2010;50:572–576. doi: 10.1016/j.neuroimage.2009.10.092. [PMC free article] [PubMed] [Cross Ref]
56. Schroeder SR, Marian V, Shook A, Bartolotti J. Bilingualism and musicianship enhance cognitive control. Neural Plast. 2016;4058620:e4058620. [PMC free article] [PubMed]

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group