Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Sep 2.
Published in final edited form as: J Neurolinguistics. 2020 Jul 14;56:100933. doi: 10.1016/j.jneuroling.2020.100933

Bilingualism modifies disengagement of attention networks across the scalp: A multivariate ERP investigation of the IOR paradigm

John G Grundy 1, Elena Pavlenko 2,3, Ellen Bialystok 4
PMCID: PMC9439621  NIHMSID: NIHMS1611630  PMID: 36061571

Abstract

A recent approach to explaining the domain-general cognitive outcomes of bilingualism is to consider the role of disengagement of attention, rather than the engagement of focused attention or inhibition as typical in most accounts. The present study pursues this approach by examining the neurophysiological changes associated with disengagement of attention in young adults performing an inhibition of return (IOR) paradigm while EEG was recorded. Participants were drawn from a diverse community and varied widely in their bilingual experience. There were three main findings. First, dividing the sample into dichotomous groups based on language proficiency did not lead to reliable group differences on the task. Second, using instead continuous measures of bilingualism across the sample indicated that greater bilingual experience and proficiency were associated with the magnitude of the IOR effect, with more bilingual individuals showing larger and earlier IOR effects. Finally, a network of processes that are temporally and spatially distinct were found to work together to produce facilitation, disengagement of attention, and inhibition of return. These findings contribute to debates regarding the electrophysiological correlates of the IOR effect and provide additional evidence for how bilingualism affects domain-general cognition.

Keywords: Bilingualism, Inhibition of Return, Disengagement of Attention, Cognition

There is substantial evidence that continual management of two languages leads to domain-general cognitive outcomes for bilinguals (review in Bialystok, 2017). The two languages are in constant competition (Kroll, Bobb, & Hoshino, 2014; Kroll, Dussias, Bogulski, & Kroff, 2012; Misra, Guo, Bobb, & Kroll, 2012) and general cognitive control mechanisms are required to select the relevant language in a given context (Abutalebi & Green, 2016; Luk, Green, Abutalebi, & Grady, 2012; Pliatsikas & Luk, 2016). Some researchers have proposed that the means of language management is through inhibition of the non-target language and that continual experience in such management leads to an enhancement of inhibitory control more broadly (Green, 1998). Evidence for this view comes from studies showing better performance by bilinguals than monolinguals on tasks that are typically considered to reflect inhibitory processes, such as the Stroop, Simon, and Flanker tasks (Bialystok et al., 2004; Costa et al., 2009; Heidlmayr et al., 2014). However, these language group effects are not always found (Kirk et al., 2014; Paap et al., 2013; Von Bastian et al., 2016), challenging both the finding of domain-general effects of bilingualism and its usual explanation based on inhibition.

A somewhat different approach is to consider the role of focusing or selective attention rather than inhibitory control as the relevant process driving these experience-related changes (Bialystok 2015, 2017). This account is better able to explain the evidence showing that infants raised in bilingual environments outperform infants from monolingual environments on tasks that involve selective attention (Comishen, Bialystok, & Adler, 2019; Kovacs & Mehler, 2009; Sebastian-Galles, Albareda-Castellot, Weikum, & Werker, 2012; Weikum et al., 2007); the infants in these studies are preverbal so inhibiting a non-target language is not a plausible explanation. The focused attention approach also helps explain why various tasks that are described as requiring inhibition, such as flanker and go-nogo, are not necessarily performed better by bilinguals; if inhibition is not the relevant process then there is no reason to expect bilinguals to excel in all tasks with this description. It also provides a better account of the evidence from language interference demonstrating that the non-target language is never actually inhibited, even in strongly monolingual contexts, and continues to influence processing (Marian & Spivey, 2003; Thierry & Wu, 2007). However, the particular attentional processes that may be affected by bilingualism remain relatively unexplored.

A recent proposal to refine the concept of attention and its possible role in bilingual processing has been to consider the role of disengagement of attention (Grundy et al., 2017; Mishra et al., 2012). This concept combines elements of both the initial view regarding inhibition in which interfering stimuli were suppressed and the later view regarding selective attention in which the target information received focus and avoids the limitations of each. Disengagement is the result of refocusing attention in response to changing task demands or cues; unlike inhibition there is no elimination of the non-target cue and like the attention view, the redirection of attention is effortful. The more efficiently this disengagement and redirection can take place, the better the performance. Applying this notion to bilingual language processing, the experience of managing two jointly-activated languages increases the efficiency of these disengagement and redirection processes even though no language representation is actually inhibited. Although there is some empirical evidence showing better disengagement by bilinguals than monolinguals for both children (Grundy & Keyvani-Chahi, 2017) and adults (Grundy et al., 20171; Mishra, Hilchey, Singh, & Klein, 2012), the neurophysiological changes associated with this disengagement have not been explored. The present study used EEG to examine how bilingualism affects the electrophysiological correlates of disengagement of attention.

The inhibition of return (IOR) paradigm (Posner & Cohen, 1984) is an ideal tool to isolate disengagement from other related attention processes because it allows one to examine the point at which attentional disengagement causes a switch from facilitating to delaying response times. The latter effect, as illustrated in Figure 1,is referred to as IOR (Klein, 2000; Klein, Castel, & Pratt, 2006).

Figure 1.

Figure 1.

Open in a new tab

The IOR paradigm used in the present study. During a typical IOR paradigm, participants are shown three horizontal boxes with a fixation cross in the central box (see Figure 1). Participants are required to make a left or a right button press indicating whether the target circle appears in the left or the right box. There are two critical manipulations. First, prior to the appearance of the target, a flash that has no predictive validity appears around the contour of either the left or the right box. If the target circle subsequently appears within the same box, it is a cued trial; if the target circle appears in the opposite box, it is an uncued trial. The second manipulation is to vary the stimulus onset asynchrony (SOA) which is the time between the flash of the cue and the appearance of the target. These two variables interact. When the target appears shortly after the flash (< 300 ms SOA), cued responses are primed by the flash and so are faster at the cued than the uncued locations; this is referred to as facilitation. In contrast, when the SOA is long (> 300 ms SOA), there is more time to disengage attention from the flash and focus back on the fixation cross. In this case, cued location responses are slower than uncued location responses because attention must return to the spot from which it just disengaged; this is known as the inhibition of return (IOR) effect. Thus, in the IOR paradigm, short SOAs lead to greater facilitation for the cued than uncued location and long SOAs lead to the IOR effect for the cued location.

Mishra et al. (2012) used the IOR paradigm to examine the hypothesis that more bilingual experience leads to more rapid disengagement of attention. Because bilinguals routinely switch between languages, interlocutors, and linguistic contexts, rapid disengagement of attention is necessary to free-up resources from the previous target so that they can be used for current task demands. Thus, practice with rapid disengagement of attention from the irrelevant language to focus on the currently relevant language could lead to an adaptation of this attention network. Mishra and colleagues examined this hypothesis by comparing a group of high proficiency bilinguals to a group of low proficiency bilinguals performing the IOR paradigm. They reported that high proficiency bilinguals switched from facilitation to IOR at earlier SOAs than low proficiency bilinguals, showing an earlier point of disengagement of attention, supporting the interpretation that more rapid disengagement of attention is associated with increasing bilingualism. These behavioral results are pursued in the present study by using EEG to investigate the electrophysiological correlates of these processes.

Previous EEG work on the IOR task has shown that disengagement is the result of early (P1) and late (N1) perceptual processes as well as later (P3) decision processes. The P1 is involved in early information processing and is known to be modulated by attention. Accordingly, several studies have shown that P1 amplitudes are smaller for cued than uncued trials during IOR (Chica and Lupiáñez, 2009; McDonald et al., 1999; Prime and Jolicoeur, 2009b; Prime and Ward, 2004; Satel et al., 2013; Tian et al., 2011; Van der Lubbe et al., 2005, Wascher and Tipper, 2004). The N1 is another perceptual process that appears immediately following the P1 and is a reflection of a sensory gating mechanism (Luck et al., 2004). The N1 has also been reported to be smaller for cued than uncued trials (McDonald et al., 1999; Tian & Yao, 2008), but this effect is less consistent (Gutiérrez-Domínguez et al., 2014). The P3 reflects stimulus categorization and response selection processes (Polich, 2007; 2012) and is sometimes larger for cued than uncued trials during the IOR (McDonald et al., 1999; Prime and Jolicoeur, 2009a). Again, however, this finding is less consistent than those for the P1 in that several studies have not reported this pattern (Chica and Lupiáñez, 2009; Hopfinger, & Mangun, 2001; Martín-Arévalo et al., 2014).

The standard approach for studies in this literature is to select components and locations of interest based on previous findings and expectations in advance and then determine whether those regions of interest conform to previous research. However, this region of interest approach may be part of the reason for the lack of consistency in this literature because there is no single electrophysiological marker for these effects (Martín-Arévalo, Chica, & Lupiáñez 2016). Instead, interactions across sites may be needed to explain these complex processes. For example, a large difference between cued and uncued trials at the P1 might lessen the need for additional processing at the subsequent N1 and P3. Similarly, if there is no difference between cued and uncued trials at the P1, an IOR effect might be explained by subsequent N1 and P3 processes. The spatial and temporal profiles of other potentially relevant processes that contribute to facilitation and IOR over time are also lacking. The present study addressed these issues by using a multivariate statistical approach called Partial Least Squares (PLS; McIntosh, Bookstein, Haxby, & Grady, 1996; Lobaugh, West, & McIntosh, 2001) on ERP data to provide an unbiased whole-brain analysis of facilitation, disengagement, and IOR processes while participants performed the IOR task. Based on previous findings, the hypothesis was that increasing levels of bilingualism will be associated with more rapid disengagement and that the P1, N1 and P3, will work together in capturing these changes in the disengagement of attention.

The present study first used the approach by Mishra et al. (2012) by dividing individuals into two groups based on proficiency in a second language, and then used recommendations in the literature to capitalize on more sensitive individual differences by using continuous measures of bilingualism (DeLuca et al., 2019; 2020; Luk, & Bialystok, 2013). EEG was recorded while participants completed the IOR task. The predictions were that: 1) greater bilingualism will lead to larger and earlier IOR effects, 2) continuous measures of bilingualism will be more sensitive than a dichotomous split in detecting these effects, and 3) facilitation, disengagement of attention, and IOR will be captured by the P1, N1, and P3 ERP components working together as a network rather than on their own.

Method

Participants and Background Measures

Forty-four participants from the undergraduate participant pool at York University in Toronto, Canada, who varied in their degree of bilingual experience participated in the experiment for course credit. The university is located in a highly diverse community where approximately 44% of people have non-English as the mother tongue and 56% of households do not use English as the primary home language; there are over 200 of these non-English languages (Statistics Canada, 2016). The community language is English and all participants were studying in a university where courses are offered in English, making English proficiency a necessary requirement. As such, their degree of bilingualism is determined by their proficiency in a non-English language, regardless of the sequence in which they learned the two languages. Furthermore, 82% of participants listed English as the first language they learned.

Thirty-six of the 44 participants listed English as the first language they learned and reported learning one of Italian (2), French (13), Igbo (1), English (5), Somali (1), Marathi (1), Gujarati (2), Cantonese (2), Punjabi (1), Hindi (1), Urdu (2), Spanish (2), Hebrew (1), or Serbian (2) as a second language. Eight of the participants listed their first language as Slovak (1), Arabic (2), Spanish (1), Punjabi (1), Taishanese (1), Mandarin (1), or Basque (1).

Participants had normal or corrected-to-normal vision and were free of any cognitive or neurological impairment, and were free of any history with head trauma. Results for the background measures are reported in Table 1.

Table 1.

Mean scores (and standard deviations) for background measures. Low and High proficiency groups were created by averaging across non-English proficiency scores in speaking, understanding, reading, and writing and then doing a median split.

Overall Low Proficiency High Proficiency
N 44 22 22
Number of males 19 10 9
Age in years 21.0 (2.5) 20.5 (2.5) 21.5 (2.5)
Shipley Vocabulary 101.5 (9.5) 102.0 (8.7) 101.0 (10.5)
Shipley Reasoning 100.2 (12.1) 98.7 (14.6) 101.6 (9.0)
SES (parents’ education /5) 3.4 (1.1) 3.2 (1.0) 3.7 (1.2)
English Proficiency (/100) 94.3 (10.5) 96.9 (6.8) 91.7 (12.8)
Non-English Proficiency (/100) 43.3 (36.8) 10.8 (12.2) 75.7 (20.5)
Non-English Usage (/4) 0.83 (0.9) 0.2 (0.4) 1.4 (0.8)
Non-English Switching (/4) 1.1 (1.2) 0.2 (0.5) 2.1 (0.9)
FS: Non-English Home 5 (2) 3.2 (0.5) 6.8 (1.0)
Use/Proficiency
FS: Non-English Social Use 5 (2) 3.9 (0.4) 6.1 (2.4)
Age of L2 Acquisition 5.9 (4.2) 9.1 (3.1) 4.0 (3.6)
Open in a new tab

Participants completed the language and social background questionnaire (LSBQ; Anderson, Mak, Chahi & Bialystok, 2018) to assess details about their bilingual experiences as well as English proficiency, age, gender, and social background information (Table 1).

Proficiency for both English and the non-English language was reported on a scale from 0 (no proficiency) to 100 (native) in each of speaking, understanding, reading, and writing. Frequency of non-English usage was assessed from 0 to 4, indicating higher degree of use of the non-English language for each of speaking, understanding, reading, and writing. Participants were also asked to estimate the age of acquisition of the language they learned second, regardless of whether it was English or Non-English. Switching frequency between languages was assessed by asking participants how often they switched between their languages with each of family, friends, and social media using a scale of 0 (never switch) to 4 (frequent switching). Factor scores from Anderson, Mak, Chahi, and Bialystok, (2018) reflecting Non-English Home Use/Proficiency and Non-English Social Use were computed from all the items in the questionnaire. For ease of interpretation, these factor scores were then converted to z-scores with a mean of 5 and standard deviation of 2.

Socioeconomic status (SES) was determined by the level of parents’ education using a 5-point scale (1= no high school diploma, 2 = high school graduate, 3 = some post-secondary education, 4 = post-secondary degree or diploma, 5 = graduate or professional degree).

The Shipley-2 Institute of Living Scale Verbal and Blocks (Shipley, Gruber, Martin, & Klein, 2009) were administered to assess English receptive vocabulary and non-verbal intelligence, respectively. Shipley measures were converted to standardized scores (μ = 100, SD = 15).

IOR Task

Figure 1 represents the IOR task used in the present manuscript.

Stimuli were presented on a 19-inch LCD screen with a black background using E-prime 2.0 (Psychology Software Tools, Inc., version 2.0.10.353). Participants sat approximately 50 cm from the screen and the refresh rate was set to 75 Hz.

Participants were asked to press “Z” for left or “M” for right on the keyboard as quickly and accurately as possible in response to a white circle that appeared for 100 ms in a left or a right box (see Figure 1) and to maintain focus on the central fixation cross throughout the experiment. Target circles appeared randomly to the left or right of the central fixation with equal probability. Prior to the appearance of the target, a cue was presented for 100 ms in the form of a thickened contour around one of the white boxes that gave the appearance of a flashing box. Participants were instructed to ignore the flash and continue responding to the position of the target circle. Following a response or 2000 ms elapsed, the fixation cross along with the three boxes was presented for 500 ms, as the inter-trial interval. For cued trials, the target appeared in the same location as the flash, and for uncued trials, the target appeared in the box opposite the flash. SOA was randomly varied between 100, 200, 300, 400, and 800 ms.

EEG recordings

A Biosemi ActiveTwo system (Amsterdam, Netherlands; www.biosemi.com) was used to acquire continuous EEG data from participants while they performed the IOR task. A total of 70 channels were used: 64 channels were placed on the scalp according to the standard 10–20 system, four eye electrodes (one below each eye and one just lateral to the outer canthi of each eye), and two electrodes placed on the right and left mastoids. The continuous signal was acquired with an open passband from DC to 150 Hz and digitized at 512 Hz. The signal was band-pass filtered off-line at 0.1–30 Hz and re-referenced to the common average reference. Offline signal processing was done using EEGLAB v11.0.2.1b and ERPLAB v5.0.0.0 toolboxes under MATLAB v7.14 (2012, Mathworks, Natick, MA). Independent components analysis (ICA) was used to identify and remove motion and eye movement artifacts from the analysis (Makeig, Bell, Jung, & Sejnowski, 1996).

Results

All analyses were performed on accuracy-adjusted RTs (RTadj), also known as efficiency scores, that are based on speed-accuracy tradeoffs (Christie & Klein, 1995; Townsend & Ashby, 1983; for a discussion on why these measures are preferred over RT or accuracy alone, see Draheim, Mashburn, Martin, & Engle, 2019). RTadj was calculated by dividing RTs by accuracy proportion so that with perfect accuracy RTs is unaffected, but as accuracy declines, RT is slowed.

Behavioral

Group analysis:

Low and high proficiency groups were created by averaging across non-English proficiency scores in speaking, understanding, reading, and writing and then doing a median split to create two groups. Low and high proficiency groups differed on all Non-English background measures (all ps < 0.001). Low proficiency and high proficiency groups did not differ on any of the other background measures, including age, F(1, 42) = 1.73, p = 0.20, gender, F < 1, SES, F(1, 42) = 2.36, p = 0.13, Shipley Vocabulary, F < 1, and Shipley Blocks, F < 1. English proficiency approached, but did not reach conventional levels of significance between groups, F(1, 42) = 3.85, p = 0.06.

RTadj by trial type and condition are plotted in Figure 2. A group (low vs. high proficiency) × SOA (100, 200, 300, 400, 800) × trial type (cued vs. uncued) mixed-measures ANOVA revealed main effects of trial type, F(1, 41) = 5.29, p = 0.03, ƞ2 = 0.11, SOA, F(4, 168) = 36.34, p < 0.001, ƞ2 = 0.46, and an interaction between them, F(4, 168) = 41.96, p < 0.001, ƞ2 = 0.50. The interaction reflects the presence of the typical IOR effect, with facilitation for cued compared to uncued trials at early SOAs, but inhibition of return at later SOAs. There were no significant effects for group or interactions between group and other variables. Therefore, the data in Figure 2 were collapsed across groups. The RTs are plotted in Panel A and the difference between cued and uncued trials are plotted in Panel B.

Figure 2.

Figure 2.

Open in a new tab

Panel A shows RTadj by trial type across SOAs for all participants. Panel B shows the cueing effects (facilitation or IOR) at each of the SOAs.

Continuous analysis:

In order to determine whether non-English descriptors (non-English proficiency, L2 age of acquisition, non-English switching frequency, non-English usage) were related to facilitation and IOR effects, separate ANCOVAs with SOA as the repeated-measure (SOA: 100, 200, 300, 400, 800) and non-English descriptor as the covariate of interest were performed on cueing effects (uncued – cued RTadj, see Figure 2B). Theoretically, a significant main effect of a covariate reveals the same information as a simple regression collapsed across SOA. However, the benefit of using an ANCOVA over a simple regression is that it allows one to examine whether the covariates of interest are related to cueing effects differently at the different SOAs.

Using non-English proficiency as the covariate of interest, there was a significant effect of SOA, F(4, 168) = 12.98, p < 0.001, ƞ2 = 0.23, and a significant effect of non-English proficiency, F(1, 42) = 4.89, p = 0.03, ƞ2 = 0.10. The main effect of proficiency demonstrates that greater non-English proficiency was associated with more likelihood of showing an IOR effect, collapsed across SOAs (r = −0.32). In other words, with greater non-English proficiency, longer RTs for cued than uncued trials were more likely. The interaction between SOA and proficiency was not significant (p > 0.6), meaning that greater non-English proficiency leads to more rapid disengagement of attention across all SOAs. No significant results were found using non-English switching frequency, non-English usage, or L2 Age of Acquisition as covariates of interest, all ps > 0.09.

For illustrative purposes in interpretation of the regression analyses presented above, Figure 3 shows how greater non-English proficiency is associated with larger IOR effects across SOAs. The figure shows the contrast between the lowest 1/3 and highest 1/3 participants according to non-English proficiency. Notice that all bars are more negative for the highest non-English proficiency (top 1/3) group compared to the lowest proficiency (bottom 1/3) group, regardless of SOA. One-sample t-tests of the cueing effects at each of the SOAs shows that those with highest non-English proficiency only show a significant facilitation effect at the 100 ms SOA, whereas those with lowest non-English proficiency show facilitation at the 100, 200, and 300 ms SOAs. Similarly, the highest non-English proficiency group shows a strong IOR effect at the 400 and 800 ms SOAs, whereas the lowest non-English proficiency group does not show a significant IOR effect at any SOA. This demonstrates the general pattern observed in the regression analyses that greater non-English proficiency leads to more likelihood of showing IOR over facilitation, and thus earlier disengagement of attention.

Figure 3.

Figure 3.

Open in a new tab

For illustrative purposes in explaining the linear regressions. Notice that all bars are more negative for the highest non-English proficiency (top 1/3) group compared to the lowest proficiency (bottom 1/3) group, regardless of SOA. *** P ≤ 0.005, ** P ≤ 0.01, * P ≤ 0.05.

Electrophysiology

To provide an unbiased whole-brain approach to identifying reliable time-windows and electrodes related to disengagement, data were analyzed using Partial Least Squares (PLS; (Lobaugh, West, & McIntosh, 2001; McIntosh, Bookstein, Haxby, & Grady, 1996). PLS is a data-driven multivariate statistical method similar to principle components analysis but constrained to the experimental conditions. Each significant latent variable (LV) that is produced represents a data-driven contrast that accounts for the greatest amount of variance in the ERP waveforms. If a second LV is significant, this represents a contrast that can explain the greatest amount of variance in the ERP waveforms after accounting for the variance explained by the first LV. One thousand permutations were computed and provided an estimate of obtaining these contrasts by chance (similar to a p-value). The electrode saliences represent the relationship between the experimental design contrasts (as represented by the LV) and the spatiotemporal pattern of ERP amplitude changes. Two hundred bootstrap re-samplings were performed to assess the reliability of electrode saliences at each time point by providing a standard error for each salience. The bootstrap procedure uses random sampling with replacement so that even though each sample will have the same number of elements as the original data, slightly different samples will be produced and reliability of the saliences can be measured. As the ratio of the salience to the standard error is approximately equal to a z-score, data points where the ratio was more than 1.7 (p < 0.05) were considered reliable. The presence of a significant LV indicates a coherent pattern of brain activation that explains most of the variance associated with the task conditions. The ERP saliences define where and when the LV is most reliable across time and space.

The PLS analysis was performed on amplitude cueing effects (uncued – cued) in order to assess the most reliable differences in ERP amplitudes that captured facilitation and IOR, using SOA (100, 200, 300, 400, 800) as the within-subjects factor. Results revealed two significant latent variables depicted in Figure 4. The first latent variable (LV1) accounted for 45.5% of the variance (p < 0.0001) and showed a pattern that mimicked the behavioral results depicted in Figure 2. Thus, this LV shows processes involved in the switch from facilitation to IOR across SOA values. Time windows were chosen based on visual inspection of the salience maps produced from the bootstrapping procedure for the PLS analysis. Figure 4 shows an example of how this can be achieved. The blue circles produced in panel C indicates that LV1 is most reliable at that location around 325–375 ms and LV2 shows reliability around 125–175 ms at that location. LV1 was reliable at multiple time windows and numerous electrode locations across the scalp, including 125–175 ms (electrodes T8, CPz, CP1, CP2, P1, Pz, Cz, C1, C6), 225–275 ms (electrodes T8, T7, TP7, F2, FC4, FC2, FCz) and 325–375 ms (electrodes T8, F1, F3, P7/T5, P07, O1, Oz).

Figure 4.

Figure 4.

Open in a new tab

PLS results for cueing effects (i.e. uncued – cued amplitudes) by SOA. In Panel A, two significant latent variables (LV1 and LV2) represent data-driven contrasts that account for the greatest amount of variance in the ERP waveforms. LV1 shows a contrast between facilitation at early SOAs and inhibition of return at later SOAs that matches the behavioral data pattern. LV2 shows a contrast between disengagement at the 300 ms SOA and facilitation and IOR at the 100 ms and 800 ms SOAs. Panel B shows where these latent variables are most reliably represented in time and space. Panel C magnifies a representative electrode for each LV showing when effects are most reliable.

Latent variable 2 (LV2) reflects disengagement processes as the LV contrasts the middle SOA of 300 ms, the typical point of disengagement, from both early and late SOAs. LV2 accounted for 31.7% of the variance (p = 0.007) after LV1 was accounted for. The pattern in LV2 is most reliable around 125–175 ms across several electrodes (Pz CPz, Fp1, AF7, F1, F3, C3, CP3, CP1, P1, P3, P5, P7, PO7, PO3, O1, Oz, POz, F2, CP4, CP2, P2, P4, P6, PO8, PO4, O2), mostly in occipital and parietal locations and a small number in left frontal regions. Figures 5 and 6 represent the temporal and spatial patterns observed from LV1 and LV2. Three temporal processes are represented in LV1, corresponding most similarly to an initial parieto-occipital P1 coupled with bi-lateral temporal P1, followed by a later fronto-central P2/N2, and a finally a left-lateralized parieto-occipital P3. As can be seen from the figure, the cueing effects at each of these sites and times contribute to the overall pattern observed indicating the switch from facilitation to IOR with a cross-over at approximately 300 ms.

Figure 5.

Figure 5.

Open in a new tab

Spatial and temporal representations of the most reliable ERPs contributing to LV1. The first set of ERPs occurring around 125–175 ms after target onset are labeled with the number “1”. The second (225–275 ms) and third (325–375 ms) subsequent ERPs are labelled “2” and “3” respectively. Blue circles represent the spatial locations of electrodes and red arrows indicate the changes across time. Waveforms shown on the right illustrate ERPs from clustered electrodes within the locations identified by LV1.

Figure 6.

Figure 6.

Open in a new tab

Spatial and temporal representations of ERP waveforms contributing to LV2. The most reliable ERPs were captured in broad posterior regions and left central regions around 125–175 ms after target onset; because they occur simultaneously, they are both labeled “1”. Blue circles represent the spatial locations of electrodes and red arrows indicate that the ERPs appeared simultaneously in both locations. Waveforms shown on the right illustrate ERPs from clustered electrodes within the locations identified by LV2.

Figure 6 represents the temporal and spatial patterns observed from LV2, corresponding most similarly to a left-lateralized frontal N1 and a broad posterior P1. Both processes work together simultaneously to distinguish facilitation and IOR from disengagement of attention.

Discussion

The present study examined the influence of bilingualism on disengagement of attention on the IOR task and identified its electrophysiological correlates. Three important findings emerged: 1) A dichotomous split based on second-language proficiency was not sufficient in the present sample to reliably detect disengagement differences between groups on the IOR task, 2) Using continuous measures of bilingualism, there was a significant relation between non-English proficiency and the magnitude and timing of the IOR effect across all SOAs, 3) A network of processes that are temporally and spatially distinct work together to produce facilitation, disengagement of attention, and inhibition of return. The findings will be discussed in turn.

Following previous work (Mishra et al., 2012), participants were first divided into high and low proficiency groups. However, in contrast to the results reported by Mishra and colleagues who found that high proficiency bilinguals switched from facilitation to IOR at an earlier SOA than low proficiency bilinguals, the present study did not reveal a significant difference between groups. In this sense, the present results initially appear to be more in line with a recent failed replication of the original study (Saint-Aubin et al., 2018). However, there are several factors that need to be considered. First, even though the Saint-Aubin et al. (2018) study did not fully replicate the Mishra et al. (2012) study, they did replicate the finding that high proficiency bilinguals were faster overall than low proficiency bilinguals. Moreover, no two groups of bilinguals can be equated as differences in linguistic factors including language switching frequency, second-language usage, and environmental contexts all place different demands on executive functions and outcomes (Beatty-Martínez et al., 2019; Bice & Kroll, 2019; Green & Abutalebi 2013; Gullifer et al., 2018; Pot, Keijzer, & De Bot, 2018). Finally, failed replications of behavioral findings need to be interpreted with caution given that fMRI and EEG studies consistently show that greater bilingualism leads to less effort on executive function tasks despite no change to the behavioral outcome (review in Grundy, Anderson, & Bialystok, 2017). All of these factors can contribute to differences in behavioral outcomes between the Mishra et al. (2012) study, the Saint-Aubin et al. (2018), and the present study. Nonetheless, group comparisons run the risk of masking real effects. This was clear from the continuous measures of bilingualism in the present study that were more revealing of disengagement and highlight the need to move away from the dichotomy and focus more on continuous individual difference measures of bilingualism (DeLuca, Rothman, Bialystok, & Pliatsikas, 2019; Luk & Bialystok, 2013). In line with this, DeLuca and colleagues (2019) argue that brain changes associated with bilingualism might be masked by the complexity of the bilingual experience. They used an individual differences approach to show that different bilingual experiences lead to separable and interactive grey and white matter adaptations.

Although the regression-based continuous analysis using the ANCOVA did not reveal an interaction with the SOA as one might initially expect, the significant effect of non-English proficiency on cueing effects across SOAs does not tell us anything about whether facilitation, IOR, or an absence of a cueing effect is observed at each of the SOAs. What is tells us is that there is less facilitation at the earlier SOAs and greater IOR effects at later SOAs, essentially pushing the cueing effects more negative (direction of IOR) equally across SOAs. This pattern is consistent with an interpretation of more likelihood of disengagement across all SOAs for those with greater non-English proficiency. This interpretation is highlighted in Figure 3, where the highest proficiency individuals show a more negative cueing effect than the lowest proficiency bilinguals across all SOAs. This (negative) push toward IOR with greater non-English proficiency leads to the observation in Figure 3 that high proficiency bilinguals only show a significant facilitation effect at the earliest 100 ms SOA, with disengagement beginning at around 200 ms, whereas disengagement begins at around 400 ms for the lowest proficiency group. These findings support the claim for an earlier switch to IOR for the high proficiency bilinguals than the low proficiency bilinguals, as reported by Mishra and colleagues at the same 200 and 400 ms SOAs, but these effects were not readily obvious until the continuous analysis was conducted.

High proficiency bilinguals had more difficulty than low proficiency bilinguals in returning to a cued than an uncued location given that they more rapidly disengaged their attention from the cue; low proficiency bilinguals were more likely to be primed by the cue. Thus, it appears that non-English proficiency does affect domain-general attentional disengagement. In line with other studies showing that bilingualism leads to enhanced attentional control (review in Bialystok, 2017), earlier and larger IOR effects are generally observed for young and healthy adult controls than young children, older adults, and people with schizophrenia, who often show no IOR effect (Klein, 2005). These findings might be explained by a prefrontal executive control deficit that does not allow these individuals to rapidly recover from a distracting cue and return to a neutral state. Additional evidence for this idea comes from a study demonstrating that the IOR effect is diminished and delayed when executive control demands are too high (Klein, Castel, & Pratt, 2006). Increasing non-English proficiency may thus enhance executive control networks that allow individuals at the highest end of the spectrum the ability to rapidly disengage attention despite high cognitive load.

Interestingly, the other language descriptor variables (age of acquisition, usage, and switching frequency) were not significantly related to attentional disengagement. Age of acquisition was not predictive of disengagement, possibly because high proficiency in non-English, the strongest predictor of disengagement, can be achieved at any age (see for example Bialystok & Kroll, 2017). It is less clear why usage and switching frequency would be less predictive of disengagement, but future studies are encouraged to explore this further.

Finally, the electrophysiological results from the present study provide the first empirical report suggesting that facilitation, disengagement of attention, and inhibition of return processes are distributed across the scalp at multiple time points and include the P1, N1, P2/N2, and P3 working together rather than on their own. Unlike previous work examining facilitation and IOR, the present study used an unbiased multivariate whole-brain approach to identify networks of processes that might be working together rather than focusing on one or two spatially and temporally constrained components. The findings presented here provide empirical evidence for the assertion that there is no single electrophysiological marker for facilitation and IOR (Martín-Arévalo, Chica, & Lupiáñez, 2016), and why there is no consensus on which component is responsible for the IOR effect: one component alone is not sufficient. The patterns of these networks were similar for participants classified as low or high bilingual proficiency, but they were activated on different time scales for participants with different degrees of non-English proficiency.

The first latent variable (LV1) in the PLS analysis almost perfectly matched the behavioral pattern observed (see Figures 3 and 5), showing a switch from facilitation to IOR. The analysis revealed that an initial parieto-occipital P1 coupled with bi-lateral temporal P1 was followed in time by a fronto-central P2/N2, and a finally a left-lateralized parieto-occipital P3. This shift for processing to begin at early parieto-occipital regions to more frontal regions and then back to occipital regions (see Figure 4) suggests that facilitation and IOR are the result of early sensory and perceptual processing of target features (P1 and N1; Mangun & Hillyard, 1991; Eimer, 1993; Han et al., 2000), followed by attention and cognitive control (P2/N2; Folstein & Van Petten, 2008), and finally stimulus categorization of the visual target (P3; Polich, 2007; 2012). The contribution of the slope between the P2 and the N2 to facilitation, disengagement, and IOR was not anticipated, but these waveforms may represent processes contributing to attentional enhancement and response selection that influence later P3 processes. Previous work has suggested that the fronto-central P2 may reflect attentional enhancement of task-relevant features (Kim et al., 2008; Potts, 2004) and the N2 may reflect response selection (Di Russo et al., 2006; Ritter, Simson, & Vaughan Jr, 1983). Source modelling places the P2 in the prefrontal cortex (Potts, 2004) and the N2 in the anterior cingulate cortex (Bekker, Kenemans, & Verbaten, 2005; Van Veen & Carter, 2002), and both of these regions play a pivotal role in task-relevant feature evaluation and response selection processes (Awh & Gehring, 1999; Bichot et al., 2015; Isomura et al., 2003), respectively. Stimulus evaluation at the P2 may affect response selection at the N2, which in turn affects processing at the P3 (Gajewski, Stoerig, & Falkenstein, 2008). The present study suggests that all three of these processes contribute to facilitation and IOR and this helps to explain why previous work has implicated the P3 but less so the P2 or the N2. What is referred to here as a P2/N2 does not contain the peak of the waveform at either the N2 or the P2. Rather, the findings point to the slopes connecting the P2 to the N2 as more reliable contributors to the IOR effect than the peaks themselves. This is important because ERP studies rarely examine differences between conditions or groups on the slopes of the waveform and assume that the peak is the most important part. Regardless, the feedback from frontal (P2/N2) to occipital (P3) regions when switching from facilitation to IOR fits with LV2 that appeared to be more directly involved in disengagement of attention.

The second latent variable (LV2) in the PLS analysis clearly distinguished the early SOA (facilitation) and late SOA (IOR) from the 300 ms SOA, where disengagement most typically appears and where the behavioral data in the present study switch from facilitation to IOR. A left-lateralized frontal N1 and a broad posterior P1 contributed to this LV. Both of these early perceptual processes appear to work together simultaneously to distinguish facilitation and IOR from disengagement of attention.

The finding that early parieto-occipital regions were largely involved in disengagement of attention is in line with the Bilingual Anterior and Subcortical Shift (BAPSS; Grundy, Anderson, & Bialystok, 2017) model and may help to explain why bilinguals show signs of dementia about 4–6 years later than monolinguals (meta-analyses in Anderson, Hawrylewicz, & Grundy, in press; Brini et al., 2020). BAPSS suggests that more second-language experience leads to less reliance on cognitively demanding frontal regions and more reliance on more automatic posterior and subcortical regions in order to be more efficient. The heavy reliance on early visual processing and the frontal and occipital coupling during disengagement is particularly in line with this view. However, despite heavy reliance on posterior processes, it is clear that a large network of processes is involved rather than a constrained set of processes to implement more rapid disengagement.

In sum, the present study demonstrated that bilingual proficiency contributes to the speed of disengagement of attention during the IOR task, that multiple processes are involved in disengagement, including early sensory and perceptual processing (P1 and N1), later attentional control (P2/N2), and visual stimulus categorization (P3). These findings contribute to our understanding of facilitation and IOR in the IOR paradigm and how linguistic experience managing two languages modifies domain-general attention networks in the brain.

Highlights.

  • More non-English proficiency leads to more rapid disengagement of attention

  • Only continuous measures reveal relationship between non-English proficiency and disengagement

  • Facilitation, disengagement of attention, and IOR are the result of multiple processes

  • Findings help to illuminate discrepancies in the field

Acknowledgments

The research reported in this paper was funded by grant R01HD052523 from the US National Institutes of Health and grant A2559 from the Natural Sciences and Engineering Research Council of Canada to EB.

Footnotes

Author statement

John G. Grundy: Conceptualization, Methodology, Formal analysis, Writing. Elena Pavlenko: Investigation, Data Curation. Ellen Bialystok: Writing, Funding acquisition.

1

See Goldsmith and Morton (2018a) for a commentary on the interpretation of our results and Bialystok and Grundy (2018) for counter-arguments. See also Goldsmith and Morton (2018b) for a “failed replication” of the original Grundy et al. (2017) findings and Grundy and Bialystok (2019) demonstrating that the study was not a replication and did not have the parameters requires to make any claims about replication.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Abutalebi J, & Green DW (2016). Neuroimaging of language control in bilinguals: Neural adaptation and reserve. Bilingualism: Language and Cognition, 19, 689–698. [Google Scholar]
  2. Anderson JAE, Hawrylewicz K, & Grundy JG (in press). Does bilingualism protect against dementia? A meta-analysis. Psychonomic Bulletin & Review. [DOI] [PubMed] [Google Scholar]
  3. Anderson JA, Mak L, Chahi AK, & Bialystok E. (2018). The language and social background questionnaire: Assessing degree of bilingualism in a diverse population. Behavior Research Methods, 50, 250–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Awh E, & Gehring WJ (1999). The anterior cingulate cortex lends a hand in response selection. Nature Neuroscience, 2, 853–854. [DOI] [PubMed] [Google Scholar]
  5. Beatty-Martínez AL., Navarro-Torres CA, Dussias PE, Bajo MT, Guzzardo Tamargo RE, & Kroll JF (2019). Interactional context mediates the consequences of bilingualism for language and cognition. Journal of Experimental Psychology: Learning, Memory, and Cognition. Advance online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bekker EM, Kenemans JL, & Verbaten MN (2005). Source analysis of the N2 in a cued Go/NoGo task. Cognitive Brain Research, 22, 221–231. [DOI] [PubMed] [Google Scholar]
  7. Bialystok E. (2015). Bilingualism and the development of executive function: The role of attention. Child Development Perspectives, 9, 117–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bialystok E. (2017). The bilingual adaptation: How minds accommodate experience. Psychological bulletin, 143, 233–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bialystok E, Craik FI, Klein R, & Viswanathan M. (2004). Bilingualism, aging, and cognitive control: evidence from the Simon task. Psychology and aging, 19, 290–303. [DOI] [PubMed] [Google Scholar]
  10. Bialystok E, & Grundy JG (2018). Science does not disengage. Cognition, 170, 330–333. [DOI] [PubMed] [Google Scholar]
  11. Bialystok E, & Kroll JF (2017). The neurobiology of language: Looking beyond monolinguals. Biolinguistics, 11, 339–351. [Google Scholar]
  12. Bice K, & Kroll JF (2019). English only? Monolinguals in linguistically diverse contexts have an edge in language learning. Brain and Language, 196, 104644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Bichot NP, Heard MT, DeGennaro EM, & Desimone R. (2015). A source for feature-based attention in the prefrontal cortex. Neuron, 88, 832–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Brini S., Sohrabi HR, Hebert JJ, Forrest MR, Laine M, Hämäläinen H, … & Fairchild TJ (2020). Bilingualism is associated with a delayed onset of dementia but not with a lower risk of developing it: a Systematic review with Meta-Analyses. Neuropsychology Review, 1–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Comishen KJ, Bialystok E, & Adler SA (2019). The impact of bilingual environments on selective attention in infancy. Developmental Science, e12797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Costa A, Hernández M, Costa-Faidella J, & Sebastián-Gallés N. (2009). On the bilingual advantage in conflict processing: Now you see it, now you don’t. Cognition, 113, 135–149. [DOI] [PubMed] [Google Scholar]
  17. Chica AB, & Lupiáñez J. (2009). Effects of endogenous and exogenous attention on visual processing: An inhibition of return study. Brain Research, 1278, 75–85. [DOI] [PubMed] [Google Scholar]
  18. Christie J, & Klein R. (1995). Familiarity and attention: Does what we know affect what we notice?. Memory & Cognition, 23, 547–550. [DOI] [PubMed] [Google Scholar]
  19. DeLuca V, Rothman J, Bialystok E, & Pliatsikas C. (2020). Duration and extent of bilingual experience modulate neurocognitive outcomes. NeuroImage, 204, 116222. [DOI] [PubMed] [Google Scholar]
  20. DeLuca V, Rothman J, Bialystok E, & Pliatsikas C. (2019). Redefining bilingualism as a spectrum of experiences that differentially affects brain structure and function. Proceedings of the National Academy of Sciences, 116, 7565–7574. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Di Russo F, Taddei F, Apnile T, & Spinelli D. (2006). Neural correlates of fast stimulus discrimination and response selection in top-level fencers. Neuroscience Letters, 408, 113–118. [DOI] [PubMed] [Google Scholar]
  22. Draheim C, Mashburn CA, Martin JD, & Engle RW (2019). Reaction time in differential and developmental research: A review and commentary on the problems and alternatives. Psychological bulletin, 145, 508. [DOI] [PubMed] [Google Scholar]
  23. Eimer M. (1993). Spatial cueing, sensory gating and selective response preparation: an ERP study on visuo-spatial orienting. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section, 88, 408–420. [DOI] [PubMed] [Google Scholar]
  24. Folstein JR., & Van Petten C. (2008). Influence of cognitive control and mismatch on the N2 component of the ERP: A review. Psychophysiology, 45, 152–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Gajewski PD, Stoerig P, & Falkenstein M. (2008). ERP—correlates of response selection in a response conflict paradigm. Brain Research, 1189, 127–134. [DOI] [PubMed] [Google Scholar]
  26. Goldsmith SF, & Morton JB (2018). Time to disengage from the bilingual advantage hypothesis. Cognition, 170, 328–329. [DOI] [PubMed] [Google Scholar]
  27. Green DW (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1, 67–81. [Google Scholar]
  28. Green DW, & Abutalebi J. (2013). Language control in bilinguals: The adaptive control hypothesis. Journal of Cognitive Psychology, 25(5), 515–530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Grundy JG (under review). The effects of bilingualism on executive functions: An updated quantitative analysis. [Google Scholar]
  30. Grundy JG, Anderson JA, & Bialystok E. (2017). Neural correlates of cognitive processing in monolinguals and bilinguals. Annals of the New York Academy of Sciences, 1396, 183–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Grundy JG, Chung-Fat-Yim A, Friesen DC, Mak L, & Bialystok E. (2017). Sequential congruency effects reveal differences in disengagement of attention for monolingual and bilingual young adults. Cognition, 163, 42–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Grundy JG, & Keyvani Chahi A. (2017). Post-conflict slowing effects in monolingual and bilingual children. Developmental science, 20, e12488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Gullifer JW., Chai XJ, Whitford V, Pivneva I, Baum S, Klein D, & Titone D. (2018). Bilingual experience and resting-state brain connectivity: Impacts of L2 age of acquisition and social diversity of language use on control networks. Neuropsychologia, 117, 123–134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Gutiérrez-Domínguez FJ, Pazo-Álvarez P, Doallo S, Fuentes LJ, Lorenzo-López L, & Amenedo E. (2014). Vertical asymmetries and inhibition of return: effects of spatial and non-spatial cueing on behavior and visual ERPs. International Journal of Psychophysiology, 91, 121–131. [DOI] [PubMed] [Google Scholar]
  35. Han S, Liu W, Yund EW, & Woods DL (2000). Interactions between spatial attention and global/local feature selection: An ERP study. Neuroreport, 11, 2753–2758 [DOI] [PubMed] [Google Scholar]
  36. Heidlmayr K, Moutier S, Hemforth B, Courtin C, Tanzmeister R, & Isel F. (2014). Successive bilingualism and executive functions: The effect of second language use on inhibitory control in a behavioural Stroop Colour Word task. Bilingualism: Language and Cognition, 17, 630–645. [Google Scholar]
  37. Hopfinger JB, & Mangun GR (2001). Tracking the influence of reflexive attention on sensory and cognitive processing. Cognitive, Affective, & Behavioral Neuroscience, 1, 56–65. [DOI] [PubMed] [Google Scholar]
  38. Isomura Y, Ito Y, Akazawa T, Nambu A, & Takada M. (2003). Neural coding of “attention for action” and “response selection” in primate anterior cingulate cortex. Journal of Neuroscience, 23, 8002–8012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kim KH, Kim JH, Yoon J, & Jung KY (2008). Influence of task difficulty on the features of event-related potential during visual oddball task. Neuroscience Letters, 445, 179–183. [DOI] [PubMed] [Google Scholar]
  40. Kirk NW, Fiala L, Scott-Brown KC, & Kempe V. (2014). No evidence for reduced Simon cost in elderly bilinguals and bidialectals. Journal of Cognitive Psychology, 26, 640–648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Klein RM (2000). Inhibition of return. Trends in Cognitive Sciences, 4, 138–147. [DOI] [PubMed] [Google Scholar]
  42. Klein RM. (2005). On the role of endogenous orienting in the inhibitory aftermath of exogenous orienting. In Mayr U, Awh E, & Keele SW (Eds.), Developing individuality in the human brain: A tribute to Michael I. Posner (pp. 45–64). Washington, DC: American Psychological Association. [Google Scholar]
  43. Klein RM, Castel AD, & Pratt J. (2006). The effects of memory load on the time course of inhibition of return. Psychonomic Bulletin & Review, 13, 294–299. [DOI] [PubMed] [Google Scholar]
  44. Kovács ÁM, & Mehler J. (2009). Cognitive gains in 7-month-old bilingual infants. Proceedings of the National Academy of Sciences, 106, 6556–6560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Kroll JF, Bobb SC, & Hoshino N. (2014). Two languages in mind: Bilingualism as a tool to investigate language, cognition, and the brain. Current Directions in Psychological Science, 23, 159–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Kroll JF, Dussias PE, Bogulski CA, & Kroff JRV (2012). Juggling two languages in one mind: What bilinguals tell us about language processing and its consequences for cognition. In Psychology of learning and motivation (Vol. 56, pp. 229–262). Academic press. [Google Scholar]
  47. Lobaugh NJ, West R, & McIntosh AR (2001). Spatiotemporal analysis of experimental differences in event-related potential data with partial least squares. Psychophysiology, 38, 517–530. [DOI] [PubMed] [Google Scholar]
  48. Luck SJ, Hillyard SA, Mouloua M, Woldorff MG, Clark VP, & Hawkins HL (1994). Effects of spatial cuing on luminance detectability: Psychophysical and electrophysiological evidence for early selection. Journal of Experimental Psychology: Human Perception and Performance, 20, 887–904. [DOI] [PubMed] [Google Scholar]
  49. Luk G, & Bialystok E. (2013). Bilingualism is not a categorical variable: Interaction between language proficiency and usage. Journal of Cognitive Psychology, 25, 605–621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Luk G., Green DW, Abutalebi J, & Grady C. (2012). Cognitive control for language switching in bilinguals: A quantitative meta-analysis of functional neuroimaging studies. Language and Cognitive rocesses, 27, 1479–1488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Makeig S, Bell AJ, Jung TP, & Sejnowski TJ (1996). Independent component analysis of electroencephalographic data. In Touretzky D, Mozer M, & Hasselmo M. (Eds.), Advances in Neural Information Processing Systems (pp. 145–151). Cambirdge, Massachusetts: The MIT Press. [Google Scholar]
  52. Mangun GR, & Hillyard SA (1991). Modulations of sensory-evoked brain potentials indicate changes in perceptual processing during visual-spatial priming. Journal of Experimental Psychology: Human Perception and Performance, 17, 1057–1074. [DOI] [PubMed] [Google Scholar]
  53. Marian V, & Spivey M. (2003). Competing activation in bilingual language processing: Within-and between-language competition. Bilingualism: Language and Cognition, 6, 97–115. [Google Scholar]
  54. Martín-Arévalo E, Chica AB, & Lupiáñez J. (2014). Electrophysiological modulations of exogenous attention by intervening events. Brain and Cognition, 85, 239–250. [DOI] [PubMed] [Google Scholar]
  55. Martín-Arévalo E, Chica AB, & Lupiáñez J. (2016). No single electrophysiological marker for facilitation and inhibition of return: A review. Behavioural Brain Research, 300, 1–10. [DOI] [PubMed] [Google Scholar]
  56. Mcdonald JJ, Ward LM, & Kiehl KA (1999). An event-related brain potential study of inhibition of return. Perception & Psychophysics, 61, 1411–1423. [DOI] [PubMed] [Google Scholar]
  57. McIntosh AR, Bookstein FL, Haxby JV, & Grady CL (1996). Spatial pattern analysis of functional brain images using partial least squares. NeuroImage, 3, 143–157. [DOI] [PubMed] [Google Scholar]
  58. Mishra RK., Hilchey MD, Singh N, & Klein RM (2012). On the time course of exogenous cueing effects in bilinguals: higher proficiency in a second language is associated with more rapid endogenous disengagement. The Quarterly Journal of Experimental Psychology, 65, 1502–1510. [DOI] [PubMed] [Google Scholar]
  59. Misra M, Guo T, Bobb SC, & Kroll JF (2012). When bilinguals choose a single word to speak: Electrophysiological evidence for inhibition of the native language. Journal of Memory and language, 67, 224–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Paap KR, & Greenberg ZI (2013). There is no coherent evidence for a bilingual advantage in executive processing. Cognitive psychology, 66, 232–258. [DOI] [PubMed] [Google Scholar]
  61. Pliatsikas C, & Luk G. (2016). Executive control in bilinguals: A concise review on fMRI studies. Bilingualism: Language and Cognition, 19, 699–705. [Google Scholar]
  62. Polich J. (2007). Updating P300: An integrative theory of P3a and P3b. Clinical Neurophysiology, 118, 2128–2148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Polich J. (2012). Neuropsychology of P300. In Kappenman ES, & Luck SJ (Eds.), Oxford handbook of event-related potentials (pp. 159–188). Oxford, UK: Oxford University Press [Google Scholar]
  64. Posner MI, & Cohen Y. (1984). Components of visual orienting. In Bouma H, Bouwhuis D. (Eds.), Attention and Performance Vol. X (pp. 531–556), Hillsdale, NJ: Erlbaum. [Google Scholar]
  65. Pot A, Keijzer M, & De Bot K. (2018). Intensity of multilingual language use predicts cognitive performance in some multilingual older adults. Brain Sciences, 8, 92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Potts GF (2004). An ERP index of task relevance evaluation of visual stimuli. Brain and Cognition, 56, 5–13. [DOI] [PubMed] [Google Scholar]
  67. Prime DJ., & Jolicœur P. (2009a). On the relationship between occipital cortex activity and inhibition of return. Psychophysiology, 46, 1278–1287. [DOI] [PubMed] [Google Scholar]
  68. Prime DJ, & Jolicoeur P. (2009b). Response-selection conflict contributes to inhibition of return. Journal of Cognitive Neuroscience, 21, 991–999. [DOI] [PubMed] [Google Scholar]
  69. Prime DJ, & Ward LM (2004). Inhibition of return from stimulus to response. Psychological Science, 15, 272–276. [DOI] [PubMed] [Google Scholar]
  70. Ritter W, Simson R, & Vaughan HG Jr (1983). Event-related potential correlates of two stages of information processing in physical and semantic discrimination tasks. Psychophysiology, 20, 168–179. [DOI] [PubMed] [Google Scholar]
  71. Saint-Aubin J, Hilchey MD, Mishra R, Singh N, Savoie D, Guitard D, & Klein RM (2018). Does the relation between the control of attention and second language proficiency generalize from India to Canada? Canadian Journal of Experimental Psychology, 72, 208–218. [DOI] [PubMed] [Google Scholar]
  72. Satel J, Hilchey MD, Wang Z, Story R, & Klein RM (2013). The effects of ignored versus foveated cues upon inhibition of return: An event-related potential study. Attention, Perception, & Psychophysics, 75, 29–40. [DOI] [PubMed] [Google Scholar]
  73. Sebastián-Gallés N, Albareda-Castellot B, Weikum WM, & Werker JF (2012). A bilingual advantage in visual language discrimination in infancy. Psychological Science, 23, 994–999. [DOI] [PubMed] [Google Scholar]
  74. Shipley WC, Gruber CP, Martin TA, & Klein AM (2009). Shipley-2. Western Psychological Services, Los Angeles, CA. [Google Scholar]
  75. Thierry G, & Wu YJ (2007). Brain potentials reveal unconscious translation during foreign-language comprehension. Proceedings of the National Academy of Sciences, 104, 12530–12535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Tian Y, Klein RM, Satel J, Xu P, & Yao D. (2011). Electrophysiological explorations of the cause and effect of inhibition of return in a cue–target paradigm. Brain Topography, 24, 164–182. [DOI] [PubMed] [Google Scholar]
  77. Tian Y, & Yao D. (2008). A study on the neural mechanism of inhibition of return by the event-related potential in the Go/Nogo task. Biological Psychology, 79, 171–178. [DOI] [PubMed] [Google Scholar]
  78. Townsend JT, & Ashby FG (1983). Stochastic modeling of elementary psychological processes. Cambridge, Massachusetts: Cambridge University Press [Google Scholar]
  79. Wascher E, & Tipper SP (2004). Revealing effects of noninformative spatial cues: An EEG study of inhibition of return. Psychophysiology, 41, 716–728. [DOI] [PubMed] [Google Scholar]
  80. Weikum WM, Vouloumanos A, Navarra J, Soto-Faraco S, Sebastián-Gallés N, & Werker JF (2007). Visual language discrimination in infancy. Science, 316, 1159–1159. [DOI] [PubMed] [Google Scholar]
  81. Van der Lubbe RHVD, Vogel RO, & Postma A. (2005). Different effects of exogenous cues in a visual detection and discrimination task: delayed attention withdrawal and/or speeded motor inhibition?. Journal of Cognitive Neuroscience, 17, 1829–1840. [DOI] [PubMed] [Google Scholar]
  82. Van Veen V, & Carter CS (2002). The anterior cingulate as a conflict monitor: fMRI and ERP studies. Physiology & Behavior, 77, 477–482. [DOI] [PubMed] [Google Scholar]
  83. Von Bastian CC, Souza AS, & Gade M. (2016). No evidence for bilingual cognitive advantages: A test of four hypotheses. Journal of Experimental Psychology: General, 145, 246–258. [DOI] [PubMed] [Google Scholar]

RESOURCES