Understanding The Consequences of Bilingualism For Language Processing and Cognition

Journal of Cognitive Psychology, 2013
http://dx.doi.org/10.1080/20445911.2013.799170
Understanding the consequences of bilingualism

for language processing and cognition
Judith F. Kroll1 and Ellen Bialystok2

1
Center for Language Science, Department of Psychology, Pennsylvania State University,
University Park, PA, USA
2
Department of Psychology, York University, Toronto, ON, Canada
Downloaded by [University of Edinburgh] at 01:09 24 June 2013
Contemporary research on bilingualism has been framed by two major discoveries. In the realm of
language processing, studies of comprehension and production show that bilinguals activate information
about both languages when using one language alone. Parallel activation of the two languages has been
demonstrated for highly proficient bilinguals as well as second language learners and appears to be
present even when distinct properties of the languages themselves might be sufficient to bias attention
towards the language in use. In the realm of cognitive processing, studies of executive function have
demonstrated a bilingual advantage, with bilinguals outperforming their monolingual counterparts on
tasks that require ignoring irrelevant information, task switching, and resolving conflict. Our claim is that
these outcomes are related and have the overall effect of changing the way that both cognitive and
linguistic processing are carried out for bilinguals. In this paper we examine each of these domains of
bilingual performance and consider the kinds of evidence needed to support this view. We argue that the
tendency to consider bilingualism as a unitary phenomenon explained in terms of simple component
processes has created a set of apparent controversies that masks the richness of the central finding in this
work: the adult mind and brain are open to experience in ways that create profound consequences for
both language and cognition.
Keywords: Bilingualism; Executive function; Language processing.
One of the most significant paradigm shifts in the isolated wire cages (Kolb et al., 2012). These
cognitive and brain sciences in the past 20 years is results simulate the well-known effects of socio-
the acceptance of the enormous potential for economic status (SES) on children’s development
plasticity at both cognitive and neuronal levels. (Farah et al., 2006). Yet there has been a
In retrospect, we should not have been so reluctance to accept that some types of pervasive
surprised: it was always known that pervasive experience could equally impact human brain
experience leaves its trace on development and structure and function. In our view, bilingualism
function. Animal studies have shown that rats is one such experience. The acquisition and use of
placed in stimulating environments that include two languages embedded in a mental conceptual
the opportunity for exercise, social interaction, structure that is at the centre of human thought
and engagement with interesting toys develop and behaviour necessarily results in a different
greater synaptic density and perform better on configuration from that found for single-language
standard maze tasks than rats kept in traditional minds. Bilingualism alters the structure and
Correspondence should be addressed to Judith F. Kroll, Center for Language Science, Department of Psychology, Pennsylvania
State University, University Park, PA 16802, USA. E-mail: [email protected]
The writing of this paper was supported in part by NIH Grant HD053146 and NSF Grants BCS-0955090 and OISE-0968369 to
JFK and by NIH Grant HD052523 and NSERC Grant A2559 to EB.
# 2013 Taylor & Francis

2 KROLL AND BIALYSTOK
function of the mind. As we will argue, bilingual vioural, imageing, and patient studies that both
minds are different not because bilingualism itself languages are active to some degree when bilin-
creates advantages or disadvantages, but because guals are using one of them (see Kroll, Dussias,
bilinguals recruit mental resources differently Bogulski, & Valdes Kroff, 2012, for a recent
from monolinguals. Those resources may be review). The evidence shows that there is a
especially critical when bilinguals comprehend bidirectional influence between languages for
and produce sentences in the less dominant of bilinguals, even in strongly monolingual contexts
their two languages, when they select the words to and even when bilinguals are highly proficient in
speak in one language only, and when they switch both languages. These effects are found whether
from one language to the other in discourse. They or not the two languages use the same writing
may also be critical during periods of development system (Hoshino & Kroll 2008; Thierry & Wu,
or decline when the networks that support lan- 2007) and even when one language is spoken
guage and cognition are potentially challenged. and the other is signed (Morford, Wilkinson,
Neuroimaging studies support this conclusion, Villwock, Piñar, & Kroll, 2011). This joint activa-
with evidence demonstrating that bilingualism tion requires a mechanism to select appropriately
changes the brain to make it more resilient and between these competing systems so that lan-
efficient in particular contexts. These effects of guage processing can proceed fluently in the
bilingualism that have been documented for target language without interference from the
language processing and for cognition more gen- other language. That mechanism is most likely
erally suggest a significant degree of adult plasti- found in the executive control system that is
city that we would not otherwise see if research largely based on a network of processes in the
were restricted to speakers of a single language. frontal cortex. Support for this interpretation
The past decade has seen an explosion in the comes from imaging studies using fMRI demon-
amount of research addressing the language and strating that the frontal executive control systems
cognitive processing of bilinguals. If we consider involved in switching between languages are the
the number of papers published and the number same as those generally used for selective atten-
of citations to research on bilingualism in the past tion to nonverbal executive function tasks (see
20 years according to Web of Science (Thompson Luk, Green, Abutalebi, & Grady, 2012, for a meta-
Reuters, 2012), there is little change between 1993 analysis) and that these executive control networks
and 2003 and then a steep and continuing rise in are used more efficiently in bilinguals than mono-
both publications and citations from that point to linguals, particularly in older bilinguals (e.g., Gold,
the present (see Figure 1). This research embo- Kim, Johnson, Kriscio, & Smith, 2013). Abutalebi
dies multiple paradigms, diverse tasks, and var- et al. (2012) identified the anterior cingulate
ious outcomes. What is clear, therefore, is cortex, a crucial part of the general executive
convergence on the idea that bilingualism is a control network, as the centre responsible
consequential life experience. What those con- for monitoring and controlling attention to two
sequences are is a matter of some debate. languages. The interpretation is also supported by
The key discovery in the research on bilingu- evidence from bilingual patients with damage to a
alism is the overwhelming evidence from beha- crucial region in the executive control system, the
Figure 1. Results of search for topic ‘‘bilingualism’’ on Thompson-Reuters Web of Science for (a) number of papers published and
(b) number of citations of those papers for years 1993 to 2012. [To view this figure in colour, please visit the online version of this
Journal.]
BILINGUAL LANGUAGE PROCESSING AND COGNITION 3
left dorsolateral prefrontal cortex, who demon- fails to capture the main point of bilingualism,
strate ‘‘pathological switching’’, the inability to namely, that it is an experience that profoundly
correctly select the intended language (Abutalebi, restructures cognitive networks and fundamen-
Miozzo, & Cappa, 2000; Fabbro, Skrap, & Aglioti, tally changes how language is processed.
2000). Thus, the mental landscape for bilinguals Two methodological issues have impeded pro-
integrates cognitive and linguistic systems in a gress in advancing our understanding of this
unique way and not surprisingly, therefore, affects problem. Although endemic to all empirical
how both cognitive and linguistic processing are research, these issues are particularly problematic
carried out. for research examining processing differences
The main findings from this body of research between monolingual and bilingual participants
on the consequences of bilingualism can all be across the lifespan because of the complexity of
traced in some measure to this joint activation of the population and the subtlety of the predicted
two language systems and nonselective access to outcomes. The first is the tendency in research to
the target system. From the perspective of lan- adopt a componential perspective in which it is
guage processing, there is evidence suggesting expected that ultimate causality can be deter-
that language comprehension and production mined for behaviour, particularly in terms of
depend on the absolute and relative levels of known constituents. The second is the assumption
proficiency of both languages, that those levels of categorical hypothesising in which it is ex-
are moderated by context and experience, and pected that mutually exclusive alternatives can be
that these processing effects are found bidirec- compared such that supporting one invalidates
tionally with each language affecting the other the other. In both cases, research on the cognitive
(e.g., Kroll & Dussias, 2013; Kroll, Dussias, et al., outcomes of bilingualism presents significant
2012). From the perspective of cognitive systems, challenges to our standard empirical approaches.
there is evidence suggesting that bilinguals at all Failing to deal adequately with the special nature
stages of the lifespan perform better than mono- of these questions will inevitably result in experi-
linguals on nonverbal executive control tasks, that ments that produce no interpretable outcomes.
bilingual performance compared to that of mono- Therefore, we begin with an examination of these
linguals depends on task materials and demands, methodological issues in terms of the specific
and that symptoms of dementia in bilinguals are problems arising from this type of research.
generally delayed relative to comparable mono-
linguals (review in Bialystok, Craik, Green, &
Gollan, 2009). What is not yet well understood is EXPLAINING THE UNKNOWN IN
how the network of cognitive resources that TERMS OF THE ‘‘KNOWN’’
regulates language processing also modifies do-
main-general cognitive and brain mechanisms; A general approach to psychological research is
that is, how does a specific experience in language to attempt to explain complex behaviour in terms
processing lead to a change in nonverbal cogni- of known components. This is an effective means
tive processing. of rendering seemingly intractable problems
Executive function advantages for bilinguals manageable by revealing their basis in simpler
have been found for tasks involving all of the processes. For example, the Stroop effect is well-
components of executive function as described by documented in cognitive psychology: The pre-
Miyake et al. (2000) and have been demonstrated sence of a printed word (i.e., a colour name)
in behavioural evidence as well as neuroimaging influences performance in a simple perceptual
using MEG (Bialystok et al., 2005) and fMRI naming task (i.e., the font colour), with facilita-
(e.g., Gold et al., 2013; Luk, Anderson, Craik, tion when the colour name and colour are
Grady, & Bialystok, 2010). Our purpose in the congruent, and interference when they are incon-
present paper is to examine the larger context in gruent. Although there are various explanations
which these linguistic and cognitive consequences for this effect, they are all grounded in models of
coexist and interact. To a great extent, research selective attention, a component of executive
has been conducted on each domain relatively processing, and within selective attention, ex-
independently, leaving the impression that bilin- plained in terms of simpler processes such as
gualism produces effects that are specific and saliency, automaticity, and parallel processing (see
isolated. In our view, this piecemeal approach for example, MacLeod & MacDonald, 2000). In
this way, an understanding of Stroop interference were presented in the centre of the screen where
is fully specified by an explanation of these com- there was no conflict and no need for inhibition.
ponent processes. Although it is a useful approach Similarly, bilinguals outperformed monolinguals
to understanding human performance on specific not only on the incongruent trials for which
tasks, our view is that this is a counterproductive inhibition was a plausible explanation, but also
and overly simplistic approach to understanding on the congruent trials where no inhibition was
the broadly based reorganisation that occurs from required. These results have been replicated in
bilingualism. Even though Stroop performance is many subsequent studies, showing that bilingualism
different for monolingual and bilingual partici- modifies not only inhibition but also monitoring
pants (Bialystok, Craik, & Luk, 2008), the cogni- (Costa, Hernández, Costa-Faidella, & Sebastian-
tive differences between monolinguals and Galles, 2009), switching (Prior & Gollan, 2011),
bilinguals are not properly captured by reference and working memory (Luo, Craik, Moreno, &
to an analytic interpretation of the Stroop effect. Bialystok, 2013; Wodniecka, Craik, Luo, &
In other words, cognitive differences between Bialystok, 2010). Moreover, bilingual advantages
monolinguals and bilinguals on the Stroop task were found for some types of inhibition but not
cannot be described as ‘‘nothing but’’ processing others; specifically, bilinguals outperformed
differences regarding saliency, automaticity, and monolinguals on tasks that required inhibition
parallel processing. There is not a one-to-one of interfering cues but not on tasks that required
mapping between bilingualism and each of these inhibition of executing a salient response (Carlson
component processes. & Meltzoff, 2008; Colzato et al., 2008; Martin-
A more general consequence of this tendency Rhee & Bialystok, 2008). Thus, a description of
is to equate tasks with the putative process most the processing components involved in tasks does
necessary to perform that task. Thus, the Stroop not correspond to the identification of processing
task becomes a ‘‘measure of inhibition’’, the n-back differences found in populations of participants.
task becomes a ‘‘measure of working memory’’, and Yet, our standard research paradigms assume that
a switching task becomes a ‘‘measure of shifting’’. such an equation is valid. For example, in a recent
Although these component processes are cer- major review of the literature on bilingualism and
tainly involved in all these tasks, they are not executive control, Hilchey and Klein (2011)
embodied by them. Task effects, group effects, assemble evidence from studies showing no
individual differences, and many other factors bilingual advantage on simple inhibition tasks
intervene between the observation that two and then use that result to discredit the entire
groups perform differently on the Stroop task body of work (see also Paap & Greenberg, 2013,
and an explanation of the cause of that difference. for a similar argument). However, as Hilchey and
This problem of attribution becomes even Klein correctly point out, there are bilingual
more difficult when the target is not a task advantages when a more holistic approach to
(‘‘Stroop task is a measure of inhibition’’) but a tasks is used. Our point is that the relations
population (‘‘bilinguals are better at inhibition’’). between complex task performance and complex
Early research on the cognitive consequences of individual characteristics cannot be reduced to
bilingualism did attempt to reduce the observa- unitary relationships.
tions to differences in known components. Thus, The inability to reduce executive function
an initial hypothesis was that bilinguals were differences between monolinguals and bilinguals
better than monolinguals at inhibiting interfer- to a single component of executive control is
ence because of their practice in inhibiting atten- consistent with emerging conceptions of this
tion to the nontarget language (e.g., Bialystok, system. Recently, Miyake and Freedman (2012)
2001). However, the limitations of this explana- acknowledged what they call the ‘‘unity and diver-
tion were apparent very early: In the first study to sity’’ of executive control, with a common core
extend the research in the cognitive advantages of shared by the component processes and unique
bilingualism to adults, Bialystok, Craik, Klein and features of different parts of the system. To
Viswanathan (2004) did indeed find bilingual accommodate this broader conception in which
advantages in a Simon task that could be attrib- there is no single cause of processing differences,
uted to inhibition, but they also reported that the researchers have begun to use more general terms
bilinguals outperformed the monolinguals in a than those given by the standard core components
condition in which participants had to press a key to explain differences in performance between
in response to one of four coloured patches that monolinguals and bilinguals. Thus, Costa et al.
(2009) argue that the bilingual advantage is in networks have developed through a different set
‘‘monitoring’’ and Bialystok (2011) attributes the of determining circumstances. Minds grow differ-
advantage to ‘‘coordination’’. It is ironic that as ently in different contexts and they grow in
we accumulate data and develop more sophisti- complex ways.
cated explanatory edifices, the explanation for
superior bilingual performance on cognitive tasks
increasingly resembles the explanation offered in
CATEGORICAL HYPOTHESES:
the first credible paper to report these effects. In
their landmark paper, Peal and Lambert (1962)
POLARISING THE ALTERNATIVES
claimed that bilinguals had greater ‘‘mental flex-
A dominant model for psychological research is
ibility’’. The search for precise components that
could ‘‘explain’’ processing differences between the orthogonal design, an approach that is based
monolinguals and bilinguals seems to have come on assessing the probability that performance
full circle. differences between tasks, conditions, or groups
The search for the correct level of description are unlikely to have occurred by chance so can be
for cognitive differences between monolinguals attributed to controlled differences between those
and bilinguals is not just a terminological issue. entities. The majority of the research on the
Assuming that tasks and groups can be ade- cognitive consequences of bilingualism has fol-
quately described in terms of single component lowed this model by comparing participants
processes leads to hypotheses that performance designated as monolingual or bilingual for their
on tasks will reveal differences between groups. ability to perform various tasks. A significant
For example, Alario, Ziegler, Massol, and De difference between groups is interpreted as evi-
Cara (2012) argued that the ability to select dence for bilingual effects on processing. How-
between nonverbal alternatives in a Simon task ever, as argued by Luk and Bialystok (this issue
should be related to the ability to select between 2013), bilingualism is not a categorical variable.
words in a naming task for monolinguals as is Approaches based on dichotomous distinctions
claimed for bilinguals (Blumenfeld & Marian, for groups or tasks, therefore, may be inappropri-
2011). They do not find such a relation and argue ate for investigations of bilingualism.
that the explanation for bilinguals is therefore The essential assumption of orthogonal designs
incorrect. This is a reductionist error: The point is is that the entities being compared are indepen-
that a more comprehensive cognitive network is dent and that variation between groups or condi-
required for bilinguals, making both linguistic and tions is confined to the variable of interest, with
cognitive processing proceed differently than they all other variables being equivalent. Bilinguals,
do for monolinguals. Reducing performance to a however, vary multidimensionally on linguistic,
few measurable components fails to capture the cognitive, social, experiential, educational, and
most crucial outcome of the experience, namely, other factors, all of which must be taken into
the reconfiguration of these networks. account when explaining performance. As such,
The tendency to argue from a simple compo- bilingualism needs to be studied in the context of
nential perspective prevents us from understand- a dynamically changing system of linguistic and
ing the linguistic and cognitive implications of cognitive performance, an approach that extends
bilingualism. Approaches based on labels applied beyond categorical assignment to groups. Relat-
to tasks and abilities that seek a correspondence edly, statistical models often assume that the
between them fail to account for the reorganisa- variables are normally distributed, a precondition
tion of whole networks that follow from bilingu- that is almost never tested yet leads to null effects
alism. The goal of current research is to identify when it is violated. Failure to obtain the gold
these correspondences and their interactions. If standard of statistical difference between groups
the new neuroscience approaches have made is often a problem of the data distribution. Finally,
anything clear, it is that there are not one-to- the overlapping distributions of two groups per-
one correspondences between the brain and forming the same task, in which participants are
behaviour. These correspondences are systematic, drawn from the same population and differ by
but not simple, for all language users. The point is only one feature, in this case bilingualism, make it
precisely that the bilingual mind comes to be extremely difficult to obtain a reliable difference
organised differently than that of the monolingual in the mean score if only one measure is being
because the representational systems and control considered. Standard experimental design usually
involves about 25 or 30 participants per group, were designated as middle class or working class
and the similarity of the populations in the two on the basis of mothers’ education. The middle-
groups, the simplicity of the tasks used in this class mothers had university degrees and the
research, and the tendency for regression towards working-class mothers had education up to and
the mean makes it astounding that significant including a high school diploma. None of the
group differences are ever obtained. The consid- families lived in poverty and in that sense, none of
erable literature that reports group differences the children were at risk for the developmental
between monolingual and bilingual participants is delays associated with those stressful environ-
greatly more informative than the attempted ments. The results showed clear effects of both
replications that fail to find significance. bilingualism and SES that were independent of
The failure to accommodate for the complex- each other and affected different behavioural
ities of bilingual experience and the limitations of outcomes. The primary effects of SES were seen
orthogonal design leads to misleading assertions. in measures of language ability and attention, and
To illustrate, some researchers have argued that the primary effects of bilingualism were seen in
group effects reported for bilingualism cannot be measures of executive functioning. Thus, as Mor-
attributed to bilingualism but instead reflect ton and Harper (2007) report, SES does indeed
differences in SES, education, immigration, or influence children’s outcomes, a point made in

culture. For example, a small-scale study by much greater detail by research studying the
Morton and Harper (2007) showed no difference effects of poverty on executive functioning (e.g.,
in performance on a Simon task between children Mezzacappa, 2004; Noble, McCandliss, & Farah,
classified as monolingual or bilingual but a sig- 2007), but such effects do not imply the absence
nificant correlation between performance and an of an effect for bilingualism. Using a different
estimate of SES. Their conclusion was that approach, Engel de Abreu, Cruz-Santos, Tourinho,
bilingualism had no effect on executive function- Martin, and Bialystok (2012) administered a large
ing (but see discussion of the tendency to over- battery of cognitive tasks to children who were all
simplify the attribution of traits to specific tasks) very low SES and carefully matched on many
because SES was the crucial variable. Extrapolat- variables. On the tasks that involved conflict and
ing from this result, they assumed that previous required executive control, bilingual children
research reporting group differences was in fact outperformed the monolinguals, but, on all other
reflecting differences in SES rather than bilingu- measures, the children in the two groups per-
alism. The error in such reasoning is to assume formed equivalently.
that categorical designs require categorical inter- The general problem with the categorical
pretations: if the effect is caused by X, then it approach is that it fails to account for the
cannot be caused by Y. inherently noncategorical nature of the relevant
The problem in applying this form of logic to constructs. Individuals are not bilingual or not,
studies of experience is that our lives are not and tasks are not measures of inhibition or not:
arranged according to a factorial design and These are all continua in which experience
multiple factors in our experience have multiple expresses itself through multiple facets and task
outcomes. All of the factors cited in the critiques complexity incorporates nuances of the domain
of the bilingualism literature (e.g., SES, educa- and the process, all of which interact in a context.
tion, etc.) do indeed influence performance on Because of the need to identify categorical
linguistic and cognitive tasks. The issue is not to variables, gradations in all of these dimensions
determine whether one of them is responsible for are washed over, and the interactions between
the outcomes but rather how they work together linguistic and cognitive systems are rarely ob-
in complex contexts and how the linguistic and served. An illustration of this problem can be
cognitive outcomes themselves interact. seen in the recent paper by Paap and Greenberg
An example of how multiple factors can be (2013) in which they call into question all
examined within the confines of a factorial design evidence for bilingual advantages in executive
comes from a recent study by Calvo (2011). function. Using young adults drawn from a highly
Multiple tests of language ability, memory, atten- heterogeneous pool of bilingual and monolingual
tion, and executive control were administered to speakers, they fail to find bilingual advantages on
5-year-old children who were monolingual or a set of behavioural tasks. Other studies have also
bilingual (all of whom had been simultaneous demonstrated that behavioural results with young
bilinguals from birth) and lived in families that adult bilinguals can sometimes be fragile and that
the very same tasks that fail to produce differences Although we have not addressed research based
for young adults may produce striking results for on neuroimaging in detail in this brief review, the
older bilinguals (e.g., Gold et al., 2013), under combination of the brain and behavioural out-
different contexts of language use (e.g., Linck, comes converge on a rather dramatic picture for
Hoshino, & Kroll, 2008), or only when cognitive the consequences of bilingualism. The primary
tasks make significant processing demands (e.g., effect appears to be that the potential to use two
Costa et al., 2009). Failures to replicate are language systems reorganises not only the pro-
important because they require that additional cesses associated with language use but also
complexity be assumed to provide a comprehen- processes involved in a number of crucial non-
sive account of the larger body of evidence. But linguistic systems (see Abutalebi & Green, 2007,
unless all conditions have been accounted for and for a review). These nonlinguistic processes,
all other explanations have been exhausted, it is particularly those associated with the executive
misleading to call into question the reliability of function system, are irrevocably altered by their
the phenomena themselves. recruitment for linguistic functions. Thus, as the
An alternative approach to investigating the bilingual mind is reconfigured to accommodate
two language systems that have different relations
consequences of bilingualism is to use multi-

variate approaches that evaluate changes on to each other, to speaker intentions, to commu-
continua. In a recent example, Bialystok and nicative contexts, and to pragmatic goals, the
Barac (2012) examined the relation between the impact of that reconfiguration is felt throughout
length of time children spent in an immersion cognitive networks. But the consequences of the
education programme and thus experienced a hypothesised accommodations will also differ for
bilingual environment, various measures of their bilinguals across the lifespan, as illustrated by the
proficiency in both languages, and outcome mea- recent Gold et al. (2013) study. The greater
sures of both nonverbal executive control and efficiency of young adult brains may make them
metalinguistic ability. The interesting finding is less likely to reveal the consequences of bilingual
that the relations were different for the different experience than their older counterparts. Not
outcomes: Bilingual experience was related to observing a bilingual effect for young adults
performance on executive control tasks and does not mean that there are no consequences
language ability was related to performance on of their language experience, but that their ability
metalinguistic tasks. Categorical comparisons be- to resolve conflict or switch effectively between
tween groups on specific tasks would not have tasks is operating within a resource-rich context
revealed these emerging differences. that has not yet been stressed by losses associated
The two methodological issues are important with normal ageing. If the apparent null effect for
considerations for all psychological research, the younger bilinguals were truly a null result,
but they are particularly challenging for studies then presumably there would not be any effect for
addressing the effect of a broad and variable the older bilinguals. Being bilingual as a young
experience, bilingualism, on the complex set of adult appears to have a cumulative effect later in
cognitive processes that comprise the executive life that produces the observed bilingual advan-
function system. Yet, in spite of these difficulties, tages in the elderly. In the Gold et al. study,
a large body of research is now identifying the the younger and older groups were so closely
consequences of bilingualism for language and matched on a host of variables that one would be
cognitive processing as revealed by both mind and hard pressed to argue that the reported effects of
brain function. bilingualism are attributable to group differences
rather than to language experience.
There is a growing interest in the role of
experience in modifying performance, an interest
LINKING THE LINGUISTIC AND that directly reflects the acceptance of plasticity
COGNITIVE SYSTEMS as a feature of minds and brains. The research
with bilingualism is in this tradition but it is
Research on the linguistic and cognitive conse- importantly different from most of the other
quences of bilingualism is just now reaching experientially based research. For most studies
maturity. As we can see in Figure 1, a critical examining the effect of experience on brain or
mass of research is accumulating and allowing us mind, the outcome of the experience is typically
to consider the full impact of this experience. in the same domain as the training and closely
related to it. Thus, string players have enhanced convergence and costs when the two languages
representation in the motor cortex responsible for diverge. Our goal in the present paper is not to
the fingers on the left hand (Elbert, Pantev, review the primary evidence for language non-
Rockstroth, Taub, & Wienbruch, 1995), jugglers selectivity because it has been discussed else-
have better visuospatial coordination than where in detail (see, for recent reviews, Kroll,
nonjugglers that is evident after brief training Bogulski, & McClain, 2012; Kroll, Dussias,
(Draganski et al., 2004), and action videogame Bogulski, & Valdes-Kroff, 2012). Rather, we focus
players have better reaction time and perceptual here on the scope of these cross-language inter-
accuracy than nongamers (Dye, Green, & actions and the consequences that the open
Bavelier, 2009). In all these cases, the training architecture of the linguistic system appears to
or experience directly results in outcomes that are create for language processing, cognition, and the
part of the performance. For bilingualism, a neural networks that support them. As in the
linguistic experience has consequences for both earlier section of the paper, our concern is with
linguistic processing, which is not entirely surpris- noncategorical aspects of processing. Just as bilin-
ing, and for nonverbal cognitive processing, which gualism is not categorical, so too different aspects
of language processing also draw on different types
is both unique to this research and unexpected.

Our explanation is that bilingualism forces lan- of resources that fall on a continuum.
guage processing to be carried out differently The notion that the bilingual cannot switch off
than it is for monolinguals, primarily because of one language at will is counterintuitive if we
joint activation of the two languages, leading to a assume that each language is represented and
reorganisation of both linguistic and cognitive processed autonomously. Superficial observation
systems. Thus, the relation between the linguistic might suggest that the bilingual is able to function
and cognitive outcomes of bilingualism is that as two monolinguals because proficient bilinguals
they are both part of the reorganisation of complex rarely make the error of using the wrong language
mental structures in response to a particular lin- (e.g., Gollan, Sandoval, & Salmon, 2011). Yet the
guistic experience. They are, in short, intimately same bilinguals may actively code switch with
interconnected and mutually interdependent. other similar speakers, switching from one lan-
The explanation in terms of broad reorganisa- guage to the other even in midsentence, and they
tion of linguistic and cognitive processes that we may have accented speech in the second language
propose is inconsistent with attempts to isolate (L2), suggesting a trace of the native language
the unique source of processing differences found and of their learning history. The data on the
for language or cognitive systems or choices parallel activation of the two languages when
between alternative categorical interpretations bilinguals are intentionally using only one of the
that are mutually exclusive. We now turn to two languages provides compelling evidence that
research on language processing to consider although they may not be aware of the activation
specifically how the consequences of bilingual of the language not in use, there is at least a
language experience might be understood within moment in processing when that information is
this framework. available (e.g., Thierry & Wu, 2007).
What is even more surprising about the emer-
ging picture of an open language system in which
there are persistent cross-language influences is
THE SCOPE AND CONSEQUENCES OF that these interactions are present for learners
CROSS-LANGUAGE ACTIVATION IN and for highly skilled bilinguals, they occur even
LANGUAGE PROCESSING when the two languages are markedly different in
form, and they are observed at every level of
The discovery that information about both lan- language processing, from the lexicon and pho-
guages is available whenever bilinguals listen to nology to the grammar. It was once thought that
speech, read, and plan spoken utterances in even these cross-language effects were more likely to
one language alone, has led to an intensive be seen in learners who have limited knowledge
examination of the scope of cross-language inter- of the L2 and for whom the more dominant first
actions. In each context of language processing, language (L1) plays a critical role in acquiring the
there is evidence that the two languages become L2 (e.g., MacWhinney, 2005). And although it is
active in parallel and compete with one another, true that learners at early stages of L2 acquisition
producing benefits when there is cross-language are particularly vulnerable to the influence of the
L1 (e.g., Sunderman & Kroll, 2006), an important READING, LISTENING, AND

discovery in this recent period of research is that SPEAKING: DIFFERENT LANGUAGE
these cross-language interactions are robust in PROCESSING TASKS
even the most proficient bilinguals. Indeed, many
of the initial studies demonstrating the activity of Although parallel activation of the bilingual’s two
the language not in use involved word recognition languages has been reported for spoken word
experiments with Dutch-English bilinguals who recognition (e.g., Marian & Spivey, 2003), visual
speak English as an L2 at a high level of pro- word recognition (Van Heuven et al., 1998), and
ficiency (e.g., Van Heuven, Dijkstra, & Grainger, spoken word production (e.g., Kroll, Bobb, &
1998). The earlier view was that the goal of L2 Wodniecka, 2006), there are differences in the
learning was to approximate the automaticity way that the coactivation of the two languages is
available in the L1 for the L2 (e.g., Segalowitz manifest across these lexical domains. As noted
& Hulstijn, 2005). The new research requires earlier, the research on visual word recognition
a revision of that position that acknowledges provides little evidence that overt cues to the
that, although L2 may indeed become more language in use modulate the activity of the
skilled with increasing proficiency, there is none- language not in use to enable language selective
theless continuing activation of both languages access. Even when the context or form of the
regardless of the level of automaticity associated word provides a clear indication of which lan-
with the L2. guage should be selected, there are persistent
For present purposes, the critical point is that effects of the language not in use that have been
cross-language activation and interaction are documented in behaviour (e.g., Schwartz, Kroll,
observed across a broad range of conditions. & Diaz, 2007), in the earliest stages of processing
Notably, it does not depend on low levels of revealed in the ERP record (e.g., Midgley,
proficiency, although proficiency affects the form Holcomb, & Grainger, 2011), and in the patterns
that these interactions take and more skilled of brain activity seen in fMRI studies (e.g., Van
bilinguals are more likely to reveal effects from Heuven & Dijkstra, 2010). For visual word
the L2 to the L1 in addition to the effects seen recognition, bilinguals simply seem not to exploit
available information to bias lexical processing to
from the L1 to the L2. At the level of lexical
the target language.
processing, there is also very little evidence that
The research on spoken word recognition
bilinguals are able to easily modulate their
generally converges with the pattern of persistent
performance to be more sensitive to the target
cross-language activation seen in visual word
language in the presence of cues that logically
recognition. For example, Lagrou, Hartsuiker,
indicate the presence of that language, such as the
and Duyck (2011) reported a series of auditory
script of a word or the language or a sentence
lexical decision experiments which included inter-
context. The effects of sentence context are lingual homophones. The typical result in the
particularly compelling because sentences pro- bilingual word recognition literature is that inter-
vide a rich source of information that goes largely lingual homophones produce interference, pre-
unused in guiding attention to the intended sumably because the parallel activation of the
language (e.g., Van Assche, Duyck, Hartsuiker, alternative meanings of the homophone across
& Diependaele, 2009). The general result in these languages results in conflict that must be resolved.
studies is that word recognition reveals cross- Lagrou et al. asked whether this pattern would be
language interactions in sentence contexts that affected by whether the spoken word was pro-
are virtually identical to those reported for word duced by a native or nonnative speaker. In theory,
recognition out of context. The only documented accented speech might provide a cue to the
exception is when sentences are highly con- language of the spoken word. They found that
strained semantically (e.g., Schwartz & Kroll, Dutch-English bilinguals were sensitive to the
2006; Van Hell & De Groot, 2008), but even difference between hearing words spoken by a
then, studies using temporally sensitive measures, native Dutch or native English speaker, but that
such as eye tracking, suggest that the locus of the sensitivity to the accentedness of speech did not
semantic constraint effect is late in processing, modulate the effect of the interlingual homo-
after cross-language activation has occurred (e.g., phone. There was similar interference for inter-
Libben & Titone, 2009). lingual homophones regardless of the speaker’s
accent. On the face of it, this result appears activated in both languages and more about
similar to the findings on bilingual word recogni- whether they compete for selection (see Kroll &
tion in the visual domain, where obvious cues do Gollan, in press, for a recent review). As in the
not effectively bias lexical access to one language studies on cognitive outcomes of bilingualism,
alone. However, other studies of spoken word the initial claim (e.g., Green, 1998) was that the
recognition have shown that under the right unintended activation of lexical candidates in
conditions, bilinguals are indeed able to exploit both of the bilingual’s two languages creates
the cues available in speech to achieve selective competition for selection that requires inhibition
access (Ju & Luce, 2004; Weber & Cutler, 2004). of the language not in use. Other models (e.g.,
It will remain for future research to identify the Costa, Miozzo, & Caramazza, 1999) argued that
conditions under which spoken word recognition language-specific selection is possible because
is language selective or not, but the evidence lexical alternatives activated in the wrong lan-
provides at least a preliminary suggestion that guage are simply not considered to be candidates
spoken word recognition may be more open to for selection. An alternative frequency-based
the influence of cues to language membership account (e.g., Gollan, Slattery, et al., 2011)
than visual word recognition. In considering the proposes that what changes in production for
mechanisms that underlie lexical selection and bilinguals relative to monolinguals is the avail-
that may subsequently affect cognition more ability of words in each language rather than
generally, these differences may turn out to be competition across the two languages. On this
as critical as the observed similarities. Cross- view, words in each of the bilingual’s two
language competition and its resolution may languages are functionally less frequent than
take different forms under different circum- words in a monolingual’s single language because
stances. Like the cognitive processes we described a bilingual has fewer opportunities to use each
earlier, it may not be a unitary phenomenon. word than a monolingual.
Perhaps the most counterintuitive evidence for It is beyond the scope of the present paper to
the parallel activation of the bilingual’s two review and contrast the evidence that has been
languages comes from research on lexical produc- taken to support each of these models of bilingual
tion, where the intention to speak a word in one production. We note that both the competition for
language is not sufficient to constrain activation selection model of production and the language-
to lexical candidates in that language (e.g., Costa, specific model allow predictions to be generated
2005; Kroll et al., 2006). In production, unlike about how a lexical selection mechanism might
word recognition, the language to be spoken must influence domain-general cognitive processes,
necessarily be selected to enable articulation. whereas the frequency model carries no obvious
Logically, then, the language of production should implications for these processes. The competition
be able to be planned in advance so that only for selection model assumes a late locus of
alternatives in that language are active. The selection, once lexical candidates have been
conceptual constraints in place to enable mean- activated in both languages; in contrast, the
ingful speech should, in theory, also be able to language-specific model assumes an early locus
encode the intended language. Yet most of the of selection, exploiting the intention to speak one
research on lexical production shows that there is language only to create a kind of mental firewall
at least momentary activation of the language not that separates the two languages. If we assume
in use. This result is perhaps not surprising when that repeated experience in language selection
bilinguals plan to speak the L2, because the L2 is creates expertise in executive function, then the
typically slower and less skilled than the L1 (e.g., competition for selection model might be hy-
Hanulovà, Davidson, & Indefrey, 2011), but it is pothesised to generate specific expertise in resol-
unexpected for planning speech in the dominant ving competition across conflicting alternatives,
L1. Moreover, the effects of cross-language com- once those alternatives are already available. In
petition in bilingual speech planning can be seen contrast, the language-specific model might be
in the earliest stages of brain activity in electro- hypothesised to affect early attentional mechan-
physiological studies for both the L1 (e.g., Misra, isms that guide the intention to establish a
Guo, Bobb, & Kroll, 2012) and the L2 (e.g., separation between alternative conditions. The
Hoshino & Thierry, 2011). frequency-based account does not provide a basis
The issue in research on bilingual production on which to generate predictions about the
has been less about whether lexical candidates are cognitive consequences of lexical selection.
For adult bilinguals, the initial evidence on the might be a single mechanism of selection that
consequences of language experience for cogni- maps onto executive control, both the require-
tion provided more support for the idea that ments for selection and the nature of the acti-
bilingualism affects later conflict resolution rather vated lexical alternatives differ for word
than early attentional biases (e.g., Hernandez, recognition and word production and may also
Costa, & Humphreys, 2012) in that the effects differ for spoken versus visual word recognition.
were mostly seen in the ability to resolve conflict In visual and spoken word recognition, lexical
between alternatives. That evidence alone, how- neighbours are the competitors, either by virtue
ever, may neither provide a sufficient basis on of similar orthography or similar phonology. In
which to adjudicate among alternative models of spoken production, semantic relatives are the
lexical selection nor an adequate account of how competitors. In each case and depending on the
bilingual experience in lexical selection comes to particular task goals in a given context, there will
alter the network that supports executive func- be cascaded activation to other lexical and sub-
tion. A number of very recent studies have lexical codes. For example, in production, the
demonstrated bilingual advantages on perceptual phonology of the planned utterance will even-
tasks that reflect early rather than late resolution
tually be activated but the planning process will

of conflict (e.g., Singh & Mishra, 2012). It is of itself be initiated at the conceptual level, with
interest to note that the evidence for crib semantic candidates activated first. As Kroll et al.
bilinguals suggests that there may indeed be an (2006) argue, the locus of selection in production
early attentional enhancement that results from may vary depending on a set of variables that
multiple language exposure early in life (e.g., include the task, the proficiency of the speaker,
Kovács & Mehler, 2009; Sebastian-Gallés, and the relative dominance of the two languages.
Albareda-Castellot, Weikum, & Werker, 2012). The implication is that there is not necessarily a
Infants appear to be exquisitely sensitive to the single locus of selection for production and,
presence of cues to each of the languages to which across production and comprehension, the locus
they are exposed. The research we have reviewed and mechanism of selection may vary. Why
on cues to language membership in bilingual word should the consequences of selecting a word in
recognition shows that adults are not as consis- one language then be the same under all of these
tently sensitive as babies to the cues that are conditions?
associated with each of their two languages, even
when those cues are quite easy to identify. The
circumstances under which adults are able to
behave more like children may turn out to be HOW CAN LANGUAGE PROCESSING
important in understanding how control in lan- HAVE CONSEQUENCES FOR
guage processing maps onto domain general COGNITION?
control processes.
In considering how cross-language activation Our discussion of cognitive differences in bilin-
and its consequences may come to affect domain- guals and language processing differences attrib-
general cognitive processes, it is also important to uted to bilingualism leaves unanswered the
note that what is activated in word production question of how these two domains are related.
differs from what is activated in word recognition. We review here three recent studies that have
Production is a conceptually driven task, engaged investigated the consequences of resolving com-
by the intention to name a picture, translate a petition in visual word recognition, spoken word
word, produce a word in response to a definition, recognition, and word production. These studies
or simply label a thought. The conceptual process illustrate the type of evidence that might allow us
that initiates speech planning requires that se- to link bilingual language processes and their
mantic alternatives are activated first and only cognitive consequences more directly. Martı́n,
later will the lexical and phonological properties Macizo, and Bajo (2010) asked Spanish-English
of the possible words be available (e.g., Levelt, bilinguals to perform semantic relatedness judge-
1989). Word recognition is fundamentally a ments to pairs of visually presented English
bottom-up process (e.g., Dijkstra & Van Heuven, words. On the critical trials of the experiment,
2002), with activation of the orthography and the word pair contained an interlingual homo-
phonology driving later semantic access. graph (e.g., pie-toe, where pie is the Spanish word
Although it may be appealing to think that there for foot) and on the following trial, the word pair
included the English translation of the Spanish In the third study, Misra et al. (2012) used a
interpretation of the homograph (e.g., foot-hand). blocked switching task in an ERP paradigm to
Bilinguals were slower to judge word pairs that investigate language selection in bilingual lexical
contained a homograph, but critically they were production. Chinese-English bilinguals named
also slower to judge a subsequent pair that pictures in blocks of trials that were either in
contained the translation of the homograph Chinese, their L1, or in English, their L2. The
rather than an unrelated control. The spillover same pictures were named in both languages. The
to the second trial was taken as an index of critical manipulation was the order in which
inhibitory control in that the English translation the languages were designated across the blocks.
had been inhibited along with attention to the The researchers predicted that naming the same
Spanish homograph on the previous trial. Martı́n pictures in the two languages would produce a
et al. found that when the two trials were pattern in the ERP data that reflected repetition
separated by 750 ms, the inhibition on the second priming; specifically, there would be reduced
trial was eliminated, suggesting that there is negativity for repeated trials where the same
momentary inhibition of the L1 interpretation picture was named in the other language. The
data supported that prediction for the L2
that is resolved within that time period.

Using a very different paradigm to test spoken (English) but the pattern for the L1 (Chinese)
word recognition, Blumenfeld and Marian (2011) was reversed. In this case, there was increased
asked bilingual and monolingual participants to negativity when the L1 followed the L2, a pattern
perform a visual world task in their L1 in which that is consistent with inhibition of the L1 after
their eye movements were monitored while they naming the pictures in L2. That negativity ex-
identified a target picture associated with a tended throughout the immediate time course of
spoken word among a display of four pictured speech planning and over many intervening trials,
even once pictures had been named in the L1
objects in a grid. One of the pictures in the display
over an entire block of trials. The use of repeated
was designed to be a phonological competitor
pictures made it impossible to identify the precise
with the named target. Both bilinguals and
scope of the observed inhibition, but the enduring
monolinguals produced the typical pattern in
inhibition over the course of the experiment
this sort of competitor task, with increased fixa-
suggests that it was long lasting, unlike the
tions to the phonological competitor relative to
inhibitory patterns observed in the word recogni-
an unrelated control. The innovation in this study
tion studies described earlier.
was to have a second nonlinguistic trial in which
These studies illustrate three points about
the same display was presented without any cross-language lexical processing. First, they pro-
pictured objects but with asterisks in the four vide evidence that bilinguals inhibit the language
picture locations; three of the asterisks were black not in use in both word recognition and word
and the target item was grey and participants production. Second, they show that the time
were asked to indicate the location of the grey course of these inhibitory processes appears to
target. Blumenfeld and Marian reported longer differ for comprehension and production. In word
reaction times interpreted as reflecting inhibition recognition, they are relatively short lived, as
for identifying the location of the target asterisk suggested by the Martı́n et al. (2010) study, but in
when it appeared in the position that had word production, they may be long lasting,
previously contained the phonological competi- persisting over many trials. Although there may
tor, but only for monolinguals and not for be shared mechanisms of inhibition, there are
bilinguals. The fact that the two groups revealed certainly distinct mechanisms as well. Future
similar patterns of phonological competition on studies will have to determine the precise scope
the first trial suggests that bilinguals were able to and time course of each of these effects. In the
resolve the inhibition of the incorrect competitor Misra et al. (2012) production study, extended
more quickly than monolinguals. Although this inhibition was not anticipated, so the study was
study did not vary the timing between the first not designed to ask the question of how long it
and second trial, the general spirit of the results is might last. Third, and perhaps most crucially, each
similar those reported by Martı́n et al. (2010), of these studies shows that it is the L1 or native
demonstrating competition in cross-language language that is inhibited. The Blumenfeld and
word recognition that appears to be resolved by Marian (2011) study was performed in the L1
inhibition. alone, yet differences were revealed between
bilingual and monolingual performance. In the early studies suggested gaps for even highly
Martı́n et al. study, the primary task was per- proficient L2 users (e.g., Johnson & Newport,
formed in the L2 but it was the L1 alternative that 1989; but see Hakuta, Bialystok, & Wiley, 2003,
was suppressed to create the observed homograph and Birdsong, 2005, for discussion of the inter-
effect. In the Misra et al. study, the act of planning pretation of critical period effects in L2 acquisi-
speech in the L2, for even relatively proficient tion). Although this issue is still under debate
bilinguals, appears to require the inhibition of the (e.g., Clahsen & Felser, 2006), the recent evidence
L1 that then has subsequent consequences for suggests greater plasticity within adult learners
speech planning in the L1. The apparent ease with than might have been predicted (e.g., Steinhauer,
which these inhibitory effects can be induced White, & Drury, 2009), much of which has been
experimentally suggests that they are available revealed by using neuroscience methods that
in natural language use. Indeed, studies of lan- track these processes more sensitively than beha-
guage when individuals are immersed in the L2 viour alone (e.g., Foucart & Frenck-Mestre, 2012).
reveal similar inhibition of the L1 (e.g., Linck, Within research on L2 acquisition, there are also
Kroll, & Sunderman, 2009). The implication is studies that examine transfer in the way that L1
knowledge may be applied to the L2 (e.g.,
that the native language changes in response to

acquiring and using a second language. The MacWhinney, 2005). Much of that work considers
accommodation that bilingualism requires of the how the presence or absence of grammatical
L1 has important implications for language pro- features in the native language affects the acquisi-
cessing itself and for traditional claims about the tion and processing of sentences in the L2 (e.g.,
privileged status of the native language. But as Tokowicz & MacWhinney, 2005).
these illustrative studies demonstrate, these But does learning and using an L2 also affect
changes within the L1 reveal one source of the the grammatical processing of the native lan-
link between language processing and their guage? Most of the research on sentence proces-
cognitive consequences. sing has compared L2 comprehension in learners
and bilinguals to native speakers of the target
language. The evidence reviewed here on bilin-
gual lexical processing suggested a high level of
DO ALL LEVELS OF BILINGUAL interaction across the two languages. These ques-
LANGUAGE PROCESSING HAVE tions have begun to be asked about the grammar
CONSEQUENCES FOR COGNITION? as well. Bilinguals typically use two languages
whose grammars are similar in some ways and
A striking feature of the research we have different in other ways. If the grammars of the
reviewed is that it is all about words, as if two languages are represented separately, then L1
language were simply a bunch of words to be sentence processing would be expected to remain
juggled and categorised. If the effects that have relatively constant as adults acquire and then
been observed for words were all we knew, they become proficient in the L2. One line of research
would be interesting, but also easy to dismiss on has addressed the influence of the L2 on L1
the grounds that the cognitive control evident in parsing preferences in Spanish-English bilinguals
bilingual language processing may be supported immersed in the L2 (e.g., Dussias, 2003; Dussias &
by the higher level context available in full Sagarra, 2007). The observation here, consistent
sentences and in actual discourse. The research with what we have seen in our review of cross-
described earlier on the effects of sentence language lexical interactions, is that the grammar
context on lexical access provides dramatic evi- of each language is influenced by the bilingual’s
dence that, at least for lexical access, a sentence experience with the other language. Dussias and
context does not appear to provide bilinguals with Sagarra (2007) asked Spanish-English bilinguals
the scaffolding needed to overcome the activity of to process sentences in Spanish that were ambig-
the language not in use. But again, that evidence uous with respect to relative clause attachment.
is about lexical processing. Spanish and English differ in their preferences for
Research on sentence processing in bilinguals high or low attachment. For example, consider the
and second language learners has focused primar- sentence: Peter fell in love with the daughter of the
ily on the question of whether late bilinguals are psychologist who studied in California. Who
capable of fully acquiring the grammar of the L2 studied in California? Native English speakers
at a level that approximates native speakers since prefer to attach low, answering that it was the
psychologist who studied in California, but native enhancements to the subcortical encoding of
Spanish speakers prefer to attach high, saying that auditory processing of speech (Krizman, Marian,
it was the daughter. Dussias and Sagarra tracked Shook, Skoe, & Kraus, 2012), to enhancing white
the eye movements of a group of highly proficient matter integrity as bilinguals age (Luk, Bialystok,
Spanish-English bilinguals living in the US while Craik, & Grady, 2011). We simply do not know
they read sentences in Spanish, their native whether learning to juggle two grammars has the
language. They found that bilinguals who had same consequences for generating these advan-
been immersed in English for a long time tages as repeatedly selecting words to speak in
processed Spanish using the English parsing one language only, or whether code switching at
preferences. These were not bilinguals who had grammatically acceptable points in midstream in
suffered language attrition, so the effect of the L2 the flow of a sentence alters the brain and
English on the L1 Spanish was not from lack of cognition in ways that differ from other sorts of
use of Spanish but rather from the effect of the multitasking.
exposure to English. Other studies using a variety From the perspective of this framework, it is
of methods have shown similar interactions across not surprising that some studies fail to reveal the
benefits of bilingualism or fail to demonstrate
the bilingual’s two languages at the level of the

syntax (e.g., Bernolet, Hartsuiker, & Pickering, effects on certain tasks. Rather, the circumstances
2007; Dussias & Cramer Scaltz, 2008; and see under which we fail to observe these conse-
Kroll & Dussias, 2013, for a recent review). Taken quences provide critical evidence to refine hy-
together, the evidence suggests that this is not just potheses about the conditions that give rise to
a story about words. Cross-language interactions them. A study by Emmorey, Luk, Pyers, and
and competition occur at the level of the grammar Bialystok (2008) illustrates the usefulness and
as well as the lexicon. What we don’t know, of also the limitations of negative findings. Emmorey
course, is whether the consequences of repeatedly et al. exploited the circumstances associated with
resolving conflict or ambiguity across two grammars bimodal bilingualism to ask whether language
draw on the same cognitive and neural mechanisms selection contributes to the bilingual advantage
that are affected by lexical competition. that has been reported on the flanker task. They
compared the flanker performance of hearing
bilinguals who use one spoken language and
another signed language, with that of hearing
HOW BILINGUALISM AFFECTS bilinguals who speak both languages, and mono-
COGNITION linguals who speak one language only. Only the
bilinguals who speak both languages showed the
In this review, we have mentioned the neu- bilingual advantage. The flanker performance for
roscience evidence only in passing. There is now the bimodal bilinguals was no better than that of
accumulating evidence for both cognitive and the monolingual controls. They reasoned that,
language measures demonstrating that bilingual- because bimodal bilinguals do not have to choose
ism has structural consequences for the brain and between their two languages to produce speech
functional consequences for neural processes. As (they can speak and cogesture at the same time,
we noted earlier, some areas of the brain, such as e.g., Emmorey, Borinstein, Thompson, & Gollan,
the anterior cingulate cortex (e.g., Abutalebi 2008), they do not have to repeatedly select the
et al., 2012) may play a particularly important language to speak. By this logic, the benefits for
role in monitoring the conflicts that characterise unimodal bilinguals can be attributed to the
bilingual language processing and its cognitive repeated requirement to select the language to
consequences. A recent paper by Zou et al. (2012) speak. Do we then assume that expertise in
shows that the brain network for the L1 is also language selection is the primary mechanism
changed by experience with the L2. What is not that affects executive control? The unimodal
yet known is how the range of language processes bilinguals in the Emmorey, Luk, et al. study
that we have reviewed impacts the documented were also faster overall, a point that has been
cognitive and neural consequences and how the discussed in the recent literature (e.g., Hilchey &
diversity of bilingual language experience mod- Klein, 2011). In addition, the bimodal bilinguals
ulates the effects that have been reported (see were children of deaf adults (CODAs) and are
Luk & Bialystok, this issue 2013). Recent reports actually heritage speakers of sign, educated and
of bilingual advantages are diverse, from dominant in their spoken language. Their failure
to produce a bilingual advantage in the flanker Bialystok, E., & Barac, R. (2012). Emerging bilingual-
task might indeed be due to the special circum- ism: Dissociating advantages for metalinguistic
awareness and executive control. Cognition, 122,
stances of bimodal bilingualism, but it may also be 67!73.
due to other aspects of their language experience. Bialystok, E., Craik, F. I. M., Grady, C., Chau, W., Ishii,
Notably, the bilinguals in the recent Zou et al. R., Gunji, A., & Pantev, C. (2005). Effect of
(2012) study that revealed a reorganisation of the bilingualism on cognitive control in the Simon task:
brain network for L1 as a result of L2 use also Evidence from MEG. NeuroImage, 24, 40!49.
Bialystok, E., Craik, F. I. M., Green, D. W., & Gollan,
tested bimodal bilinguals who used one spoken T. H. (2009). Bilingual minds. Psychological Science
and one signed language. Bimodal bilingualism in the Public Interest, 10, 89!129.
may have some similar and some different con- Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan,
sequences relative to unimodal bilingualism. M. (2004). Bilingualism, aging, and cognitive con-
The goal of this paper was to examine the story trol: Evidence from the Simon task. Psychology and
Aging, 19, 290!303.
that we and others have told in the past about Bialystok, E., Craik, F. I. M., & Luk, G. (2008). Cognitive
bilingualism and its cognitive and linguistic con- control and lexical access in younger and older
sequences. The evidence that we have reviewed bilinguals. Journal of Experimental Psychology:
and the rapidly emerging findings on the con- Learning, Memory, and Cognition, 34, 859!873.
sequences of bilingualism make clear that the Birdsong, D. (2005). Interpreting age effects in second
language acquisition. In J. F. Kroll & A. M. B. De
bilingual is indeed a mental juggler at all levels of Groot (Eds.), Handbook of bilingualism: Psycholin-
language processing and that there are a host of guistic approaches (pp. 109!127). New York, NY:
consequences that result, many of which can be Oxford University Press.
characterised as benefits. But which aspects of Blumenfeld, H. K., & Marian, V. (2011). Bilingualism
that juggling produce which constellation of influences inhibitory control in auditory comprehen-
sion. Cognition, 118, 245!257.
consequences are just beginning to emerge. The Calvo, A. (2011). The role of bilingualism and socio-
resulting research will not only enhance our economic status in executive functioning (Unpublished
understanding of how language experience affects dissertation). York University, Toronto, Canada.
the mind and the brain, but will also require Carlson, S. M., & Meltzoff, A. N. (2008). Bilingual
nothing short of a revision to traditional accounts experience and executive functioning in young
children. Developmental Science, 11, 282!298.
of language development. Clahsen, H., & Felser, C. (2006). Grammatical proces-
sing in language learners. Applied Psycholinguistics,
27, 3!42.
Colzato, L. S., Bajo, M. T., van den Wildenberg, W.,
REFERENCES Paolieri, D., Nieuwenhuis, S., La Heij, W., &
Hommel, B. (2008). How does bilingualism improve
Abutalebi, J., Della Rosa, P. A., Green, D. W., executive control? A comparison of active and
Hernandez, M., Scifo, P., Keim, R., . . . Costa, A. reactive inhibition mechanisms. Journal of Experi-
(2012). Bilingualism tunes the anterior cingulate mental Psychology: Learning, Memory, and Cogni-
cortex for conflict monitoring. Cerebral Cortex, 22, tion, 34, 302!312.
2076!2086. Costa, A. (2005). Lexical access in bilingual production.
Abutalebi, J., Miozzo, A., & Cappa, S. F. (2000). Do In J. F. Kroll & A. M. B. De Groot (Eds.), Hand-
subcortical structures control ‘language selection’ in book of bilingualism: Psycholinguistic approaches
polyglots? Evidence from pathological language (pp. 308!325). New York, NY: Oxford University
mixing. Neurocase, 6, 51!56. Press.
Alario, F.-X., Ziegler, J. C., Massol, S., & De Cara, B. Costa, A., Hernández, M., Costa-Faidella, J., &
(2012). Probing the link between cognitive control Sebastian-Galles, N. (2009). On the bilingual advan-
and lexical selection in monolingual speakers. tage in conflict processing: Now you see it, now you
L’Année Psychologique/Topics in Cognitive Psychol- don’t. Cognition, 113, 135!149.
ogy, 112, 545!559. Costa, A., Miozzo, M., & Caramazza, A. (1999).
Bernolet, S., Hartsuiker, R. J., & Pickering, M. J. Lexical selection in bilinguals: Do words in the
(2007). Shared syntactic representations in bilin- bilingual’s two lexicons compete for selection?
guals: Evidence for the role of word-order repeti- Journal of Memory and Language, 41, 365!397.
tion. Journal of Experimental Psychology: Learning, Dijkstra, T., & Van Heuven, W. J. B. (2002). The
Memory, and Cognition, 33, 931!949. architecture of the bilingual word recognition sys-
Bialystok, E. (2001). Bilingualism in development: tem: From identification to decision. Bilingualism:
Language, literacy, and cognition. New York, NY: Language and Cognition, 5, 175!197.
Cambridge University Press. Draganski, B., Gaser, C., Busch, V., Schuierer, G.,
Bialystok, E. (2011). Coordination of executive func- Bogdahn, U., & May, A. (2004). Neuroplasticity:
tions in monolingual and bilingual children. Journal Changes in grey matter induced by training. Nature,
of Experimental Child Psychology, 110, 461!468. 427, 311!312.
Dussias, P. E. (2003). Syntactic ambiguity resolution in second language acquisition. Psychological Science,
L2 learners: Some effects of bilinguality on LI and 14, 31!38.
L2 processing strategies. Studies in Second Hanulovà, J., Davidson, D. J., & Indefrey, P. (2011).
Language Acquisition, 25, 529!557. Where does the delay in L2 picture naming come
Dussias, P. E., & Cramer Scaltz, T. R. (2008). Spanish- from? Psycholinguistic and neurocognitive evidence
English L2 speakers’ use of subcategorization bias on second language word production. Language and
information in the resolution of temporary ambi- Cognitive Processes, 26, 902!934.
guity during second language reading. Acta Psycho- Hernández, M., Costa, A., & Humphreys, G. W. (2012).
logica, 128, 501!513. Escaping capture: Bilingualism modulates distrac-
Dussias, P. E., & Sagarra, N. (2007). The effect of tion from working memory. Cognition, 122, 37!50.
exposure on syntactic parsing in Spanish-English Hilchey, M. D., & Klein, R. M. (2011). Are there
bilinguals. Bilingualism: Language and Cognition, bilingual advantages on nonlinguistic interference
10, 101!116. tasks? Implications for the plasticity of executive
Dye, M., Green, W. G., & Bavelier, D. (2009). Increas- control processes. Psychonomic Bulletin and Review,
ing speed of processing with action video games. 18, 625!658.
Current Directions in Psychological Science, 18, Hoshino, N., & Kroll, J. F. (2008). Cognate effects in
321!326. picture naming: Does cross-language activation
Elbert, T., Pantev, C., Rockstroth, B., Taub, E., & survive a change of script? Cognition, 106, 501!511.
Hoshino, N., & Thierry, G. (2011). Language selection
Wienbruch, C. (1995). Increased cortical representa-

tion of the fingers of the left hand in string players. in bilingual word production: Electrophysiological
Science, 270, 305!307. evidence for cross-language competition. Brain
Emmorey, K., Borinstein, H. B., Thompson, R. L., & Research, 1371, 100!109.
Gollan, T. H. (2008). Bimodal bilingualism. Bilingu- Johnson, J. S., & Newport, E. L. (1989). Critical period
alism: Language and Cognition, 11, 43!61. effects in second language learning: The influence of
Emmorey, K., Luk, G., Pyers, J. E., & Bialystok, E. maturational state on the acquisition of English as a
(2008). The source of enhanced cognitive control in second language. Cognitive Psychology, 21, 60!99.
bilinguals. Psychological Science, 19, 1201!1206. Ju, M., & Luce, P. A. (2004). Falling on sensitive ears:
Engel de Abreu, P. M. J., Cruz-Santos, A., Tourinho, C. Constraints on bilingual lexical activation. Psycho-
J., Martin, R., & Bialystok, E. (2012). Bilingualism logical Science, 15, 314!318.
enriches the poor: Enhanced cognitive control in Kolb, B., Mychasiuk, R., Muhammad, A., Li, Y., Frost,
low-income minority children. Psychological D. O., & Gibb, R. (2012). Experience and the
Science, 23, 1364!1371. developing prefrontal cortex. Proceedings of the
Fabbro, F., Skrap, M., & Aglioti, S. (2000). Pathological National Academy of Sciences, 109, 17186!17193.
switching between languages after frontal lesions in Kovács, A. M., & Mehler, J. (2009). Cognitive gains in
a bilingual patient. Journal of Neurology, Neurosur- 7-month-old bilingual infants. Proceedings of the
gery and Psychiatry, 68, 650!652. National Academy of Sciences, 106, 6556!6560.
Farah, M. J., Shera, D. M., Savage, J. H., Betancourt, L., Krizman, J., Marian, V., Shook, A., Skoe, E., & Kraus,
Giannetta, J. M., Brodsky, N. L., . . . Hurt, H. (2006). N. (2012). Subcortical encoding of sound is en-
Childhood poverty: Specific associations with neu- hanced in bilinguals and relates to executive func-
rocognitive development. Brain Research, 1110, tion advantages. Proceedings of the National
166!174. Academy of Sciences, 109, 7877!7881.
Foucart, A., & Frenck-Mestre, C. (2012). Can late L2 Kroll, J. F., Bobb, S., & Wodniecka, Z. (2006).
learners acquire new grammatical features? Evi- Language selectivity is the exception, not the rule:
dence from ERPs and eye-tracking. Journal of Arguments against a fixed locus of language selec-
Memory and Language, 66, 226!248. tion in bilingual speech. Bilingualism: Language and
Gold, B. T., Kim, C., Johnson, N. F., Kriscio, R. J., & Cognition, 9, 119!135.
Smith, C. D. (2013). Lifelong bilingualism maintains Kroll, J. F., Bogulski, C. A., & McClain, R. (2012).
neural efficiency for cognitive control in aging. Psycholinguistic perspectives on second language
Journal of Neuroscience, 33, 387!396. learning and bilingualism: The course and conse-
Gollan, T. H., Sandoval, T., & Salmon, D. P. (2011). quence of cross-language competition. Linguistic
Cross-language intrusion errors in aging bilinguals Approaches to Bilingualism, 2, 1!24.
reveal the link between executive control and Kroll, J. F., & Dussias, P. E. (2013). The comprehension
language selection. Psychological Science, 22, of words and sentences in two languages. In
1155!1164. T. Bhatia & W. Ritchie (Eds.), The handbook of
Gollan, T. H., Slattery, T. J., Goldenberg, D., Van bilingualism and multilingualism (2nd ed., pp. 216!
Assche, E., Duyck, W., & Rayner, K. (2011). 243). Malden, MA: Wiley-Blackwell.
Frequency drives lexical access in reading but not Kroll, J. F., Dussias, P. E., Bogulski, C. A., & Valdes-
in speaking: The frequency-lag hypothesis. Journal Kroff, J. (2012). Juggling two languages in one mind:
of Experimental Psychology: General, 140, 186!209. What bilinguals tell us about language processing
Green, D. (1998). Mental control of the bilingual and its consequences for cognition. In B. Ross (Ed.),
lexico-semantic system. Bilingualism: Language The psychology of learning and motivation (Vol. 56,
and Cognition, 1, 67!81. pp. 229!262). San Diego, CA: Academic Press.
Hakuta, K., Bialystok, E., & Wiley, E. (2003). Critical Kroll, J. F., & Gollan, T. H. (in press). Speech planning
evidence: A test of the critical period hypothesis for in two languages: What bilinguals tell us about
language production. In V. Ferreira, M. Goldrick, & sociodemographic correlates in an epidemiological

M. Miozzo (Eds.), The Oxford handbook of lan- sample of young, urban children. Child Develop-
guage production. Oxford: Oxford University Press. ment, 75, 1373!1386.
Lagrou, E., Hartsuiker, R. J., & Duyck, W. (2011). Midgley, K. J., Holcomb, P. J., & Grainger, J. (2011).
Knowledge of a second language influences auditory Effects of cognate status on word comprehension in
word recognition in the native language. Journal of second language learners: An ERP investigation.
Experimental Psychology: Learning, Memory, and Journal of Cognitive Neuroscience, 23, 1634!1647.
Cognition, 37, 952!965. Misra, M., Guo, T., Bobb, S. C., & Kroll, J. F. (2012).
Levelt, W. J. M. (1989). Speaking: From intention to When bilinguals choose a single word to speak:
articulation. Cambridge, MA: MIT Press. Electrophysiological evidence for inhibition of the
Libben, M. R., & Titone, D. A. (2009). Bilingual lexical native language. Journal of Memory and Language,
access in context: Evidence from eye movements 67, 224!237.
during reading. Journal of Experimental Psychology: Miyake, A., & Friedman, N. P. (2012). The nature and
Learning, Memory, and Cognition, 35, 381!390. organization of individual differences in executive
Linck, J. A., Hoshino, N., & Kroll, J. F. (2008). Cross- functions: Four general conclusions. Current Direc-
language lexical processes and inhibitory control. tions in Psychological Science, 21, 8!14.
Mental Lexicon, 3, 349!374. Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A.
Linck, J. A., Kroll, J. F., & Sunderman, G. (2009). Losing H., Howerter, A., & Wager, T. D. (2000). The unity
access to the native language while immersed in a
and diversity of executive functions and their con-

second language: Evidence for the role of inhibition tributions to complex ‘‘frontal lobe’’ tasks: A latent
in second language learning. Psychological Science, variable analysis. Cognitive Psychology, 41, 49!100.
20, 1507!1515. Morford, J. P., Wilkinson, E., Villwock, A., Piñar, P., &
Luk, G., Anderson, J. A. E., Craik, F. I. M., Grady, C., Kroll, J. F. (2011). When deaf signers read English:
& Bialystok, E. (2010). Distinct neural correlates for Do written words activate their sign translations?
two types of inhibition in bilinguals: Response Cognition, 118, 286!292.
inhibition versus interference suppression. Brain Morton, J. P., & Harper, S. N. (2007). What did Simon
and Cognition, 74, 347!357. say? Revisiting the bilingual advantage. Develop-
Luk, G., & Bialystok, E. (2013). Bilingualism is not a mental Science, 10, 719!726.
categorical variable. Journal of Cognitive Psychology. Noble, K. G., McCandliss, B. D., & Farah, M. J. (2007).
Luk, G., Bialystok, E., Craik, F. I. M., & Grady, C. Socioeconomic gradients predict individual differ-
(2011). Lifelong bilingualism maintains white matter ences in neurocognitive abilities. Developmental
integrity in older adults. Journal of Neuroscience, 31, Science, 10, 464!480.
16808!16813. Paap, K. R., & Greenberg, Z. I. (2013). There is no
Luk, G., Green, D. W., Abutalebi, J., & Grady, C. coherent evidence for a bilingual advantage in
(2012). Cognitive control for language switching in executive processing. Cognitive Psychology, 66,
bilinguals: A quantitative meta-analysis of func- 232!258.
tional neuroimaging studies. Language and Cogni- Peal, E., & Lambert, W. (1962). The relation of
tive Processes, 27, 1479!1488. bilingualism to intelligence. Psychological Mono-
Luo, L., Craik, F. I. M., Moreno, S., & Bialystok, E. graphs: General and Applied, 76, 1!23.
(2013). Bilingualism interacts with domain in a Prior, A., & Gollan, T. H. (2011). Good language-
working memory task: Evidence from aging. Psy- switchers are good task-switchers: Evidence from
chology and Aging, 28, 28!34. Spanish-English and Mandarin-English bilinguals.
MacLeod, C. M., & MacDonald, P. A. (2000). Inter- Journal of the International Neuropsychological
dimensional interference in the Stroop effect: Un- Society, 17, 682!691.
covering the cognitive and neural anatomy of Schwartz, A. I., & Kroll, J. F. (2006). Bilingual lexical
attention. Trends in Cognitive Sciences, 4, 383!391. activation in sentence context. Journal of Memory
MacWhinney, B. (2005). A unified model of language and Language, 55, 197!212.
acquisition. In J. F. Kroll & A. M. B. De Groot Schwartz, A., Kroll, J. F., & Diaz, M. (2007). Reading
(Eds.), Handbook of bilingualism: Psycholinguistic words in Spanish and English: Mapping orthography
approaches (pp. 49!67). New York, NY: Oxford to phonology in two languages. Language and
University Press. Cognitive Processes, 22, 106!129.
Marian, V., & Spivey, M. J. (2003). Competing activa- Sebastián-Gallés, N., Albareda-Castellot, B., Weikum,
tion in bilingual language processing: Within- and W. M., & Werker, J. F. (2012). A bilingual advantage
between-language competition. Bilingualism: Lan- in visual language discrimination in infancy. Psycho-
guage and Cognition, 6, 97!115. logical Science, 23, 994!999.
Martı́n, M. C., Macizo, P., & Bajo, T. (2010). Time course Segalowitz, N., & Hulstijn, J. (2005). Automaticity in
of inhibitory processes in bilingual language proces- bilingualism and second language learning. In J. F.
sing. British Journal of Psychology, 101, 679!693. Kroll & A. M. B. De Groot (Eds.), Handbook of
Martin-Rhee, M. M., & Bialystok, E. (2008). The bilingualism: Psycholinguistic approaches (pp. 371!
development of two types of inhibitory control in 388). New York, NY: Oxford University Press.
monolingual and bilingual children. Bilingualism: Singh, N., & Mishra, R. K. (2012). Does language
Language and Cognition, 11, 81!93. proficiency modulate oculomotor control? Evidence
Mezzacappa, E. (2004). Alerting, orienting, and from Hindi-English bilinguals. Bilingualism: Lan-
executive attention: Developmental properties and guage and Cognition, 15, 771!781.
Steinhauer, K., White, E. J., & Drury, J. E. (2009). Van Hell, J. G., & De Groot, A. M. B. (2008). Sentence
Temporal dynamics of late second language acquisi- context modulates visual word recognition and trans-
tion: Evidence from event-related brain potentials. lation in bilinguals. Acta Psychologica, 128, 431!451.
Second Language Research, 25, 13!41. Van Heuven, W. J. B., & Dijkstra, T. (2010). Language
Sunderman, G., & Kroll, J. F. (2006). First language comprehension in the bilingual brain: fMRI and
activation during second language lexical proces- ERP support for psycholinguistic models. Brain
sing: An investigation of lexical form, meaning, and Research Reviews, 64, 104!122.
grammatical class. Studies in Second Language Van Heuven, W. J. B., Dijkstra, T., & Grainger, J.
Acquisition, 28, 387!422. (1998). Orthographic neighborhood effects in bilin-
Thierry, G., & Wu, Y. J. (2007). Brain potentials reveal gual word recognition. Journal of Memory and
unconscious translation during foreign-language Language, 39, 458!483.
comprehension. Proceedings of the National Acad- Weber, A., & Cutler, A. (2004). Lexical competition in
emy of Sciences, 104, 12530!12535. non-native spoken-word recognition. Journal of
Thompson-Reuters. (2012). Web of science. Retrieved 1 Memory and Language, 50, 1!25.
October 2012 from http://www.isiknowledge.com Wodniecka, Z., Craik, F. I. M., Luo, L., & Bialystok, E.
Tokowicz, N., & MacWhinney, B. (2005). Implicit and (2010). Does bilingualism help memory? Competing
explicit measures of sensitivity to violations in effects of verbal ability and executive control.
second language grammar: An event-related poten- International Journal of Bilingual Education and
tial investigation. Studies in Second Language Ac- Bilingualism, 13, 575!595.
quisition, 27, 173!204. Zou, L., Abutalebi, J., Zinszer, B., Yan, X., Shu, H.,
Van Assche, E., Duyck, W., Hartsuiker, R. J., & Peng, D., & Ding, G. (2012). Second language
Diependaele, K. (2009). Does bilingualism change experience modulates functional brain network for
native-language reading? Cognate effects in a sen- the native language production in bimodal bilin-
tence context. Psychological Science, 20, 923!927. guals. NeuroImage, 62, 1367!1375.

Understanding The Consequences of Bilingualism For Language Processing and Cognition

Uploaded by

Copyright:

Available Formats

Understanding The Consequences of Bilingualism For Language Processing and Cognition

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Understanding The Consequences of Bilingualism For Language Processing and Cognition

Uploaded by

Copyright:

Available Formats

Journal of Cognitive Psychology, 2013

Understanding the consequences of bilingualism

Judith F. Kroll1 and Ellen Bialystok2

Keywords: Bilingualism; Executive function; Language processing.

# 2013 Taylor & Francis

differences in SES, education, immigration, or influence children’s outcomes, a point made in

consequences of bilingualism is to use multi-

is both unique to this research and unexpected.

L1 (e.g., Sunderman & Kroll, 2006), an important READING, LISTENING, AND

tually be activated but the planning process will

that is resolved within that time period.

that the native language changes in response to

the bilingual’s two languages at the level of the

Wienbruch, C. (1995). Increased cortical representa-

language production. In V. Ferreira, M. Goldrick, & sociodemographic correlates in an epidemiological

and diversity of executive functions and their con-

You might also like