How Does The Bilingual Experience Sculpt The Brain?

PERSPECTIVES
information (such as the location of articles

OPINION
and prepositions) of the language, among
many other features. For example, infants
How does the bilingual experience learning Japanese have to learn that arti-
cles and prepositions come after nouns,
sculpt the brain? whereas infants learning English or Spanish
have to learn the opposite. The develop-
mental trajectory of acquiring a language
Albert Costa and Núria Sebastián-Gallés in a monolingual environment is relatively
well described3,4, but much less is known
Abstract | The ability to speak two languages often marvels monolinguals, although about language learning in infants raised in
bilinguals report no difficulties in achieving this feat. Here, we examine how a bilingual (or multilingual) environment.
learning and using two languages affect language acquisition and processing as Nevertheless, as we argue, linguistic develop-
well as various aspects of cognition. We do so by addressing three main questions. ment in monolingual and bilingual contexts
First, how do infants who are exposed to two languages acquire them without is similar despite the fact that the ‘bilingual
experience’ is associated with some specific
apparent difficulty? Second, how does language processing differ between adaptations in the learning process.
monolingual and bilingual adults? Last, what are the collateral effects of There are two main differences between
bilingualism on the executive control system across the lifespan? Research in all acquisition of a language in a monolingual
three areas has not only provided some fascinating insights into bilingualism but context and acquisition of two languages in
also revealed new issues related to brain plasticity and language learning. a simultaneous bilingual context. The first
of these is quantitative, as bilingual infants
must learn two linguistic codes instead of
The pervasive presence of bilingualism successive bilinguals) (BOX 1) and the effects just one (that is, two sets of phonemes, two
shows that humans can learn two languages of bilingualism on the mechanisms of cog- lexicons and two grammatical systems).
without apparent difficulty. However, bilin- nitive functioning outside the language Furthermore, presumably, learning both
gualism is difficult to define, as it encom- domain. These are the main topics of the codes needs to be achieved in the context
passes a broad typology of speakers. Indeed, present article. Note that we only briefly of reduced exposure to each of the two lan-
the acquisition of two languages may occur cover issues related to learning a second languages, as there is no reason to assume that,
in many different contexts. People can learn guage and the brain representation of two overall, bilingual parents speak more to their
two languages from birth (such individu- languages (see REFS 2–4 for reviews of these children than do monolingual parents.
als are known as simultaneous bilinguals topics and BOX 3). The second difference is qualitative and
(BOX 1), as both languages are learned at the In the first section of this Opinion article, relates to bilinguals’ requirement to perform
same time). Alternatively, they can learn a we focus on the language-learning challenges specific computations that monolinguals
second language later in life under formal faced by infants who are exposed to two lan- do not have to perform. Bilingual infants
instruction, in an immersion environ- guages from the beginning of life (sometimes need to be able to notice the existence of
ment as a result of immigration or in one called ‘bilingual first-language acquisition’ more than one ‘type of speech’ and, then, to
of many other situations (such individuals (REF. 2)). The studies show that although the adequately sort and parse the information
are known as successive bilinguals (BOX 1)). pattern of development is not fundamentally corresponding to each of the speech types.
In addition, among bilinguals, individuals different in bilingual versus monolingual Thus, the learning of two (or more) language
can show considerable variation in the pro- populations, the bilingual input induces some systems runs in parallel with the need to sort
ficiency levels attained in their languages: specific learning adaptations. In the second and properly compute the information for
whereas some show equal proficiency in section, we review studies that have addressed each language. These important quantitative
both languages, others have a clear domi- how learning a second language affects first- and qualitative differences between mono-
nance in one of their languages1. These language comprehension and production lingual and bilingual learning contexts need
differences in learning contexts and profi- processes. These studies have mostly focused to be considered when analysing language
ciency level pose methodological challenges on young adult successive bilinguals. Last, learning (BOX 4).
in bilingual studies (BOX 2). in the third section, we discuss the potential
Most bilingual research has been carried collateral effects of bilingualism on domain- Language discrimination. A potential chal-
out on adult successive bilinguals, specifi- general executive control processes. The stud- lenge faced by bilingual infants is the need to
cally focusing on second-language process- ies discussed in this section take a life-long discriminate between the two languages that
ing in the brain and how it is affected by perspective, examining simultaneous and they are exposed to. A failure to discriminate
the age at which the second language is successive bilingual infants, children, young may cause difficulties in ‘cracking’ the lin-
acquired and the amount of exposure an adults and the elderly. guistic codes of the two languages. However,
individual has to that language. However, language discrimination does not seem to
the study of bilingualism should also Language acquisition in bilingual infants pose a major difficulty for infants. Several
examine how simultaneously learning two Learning a language involves acquiring studies have shown that, at birth, humans pre-
languages affects the development of these knowledge about the specific properties natally exposed to monolingual or bilingual
languages (in simultaneous bilinguals), of that language. Infants have to learn the inputs are able to differentiate between two
the impact of learning a second language specific phoneme repertoire (BOX 1), the languages, provided they sound very differ-
on the processing of the first language (in words and the sophisticated grammatical ent, such as Tagalog and English5 or Dutch
336 | MAY 2014 | VOLUME 15 www.nature.com/reviews/neuro
© 2014 Macmillan Publishers Limited. All rights reserved

PERSPECTIVES
Box 1 | Key terms and concepts

Cognitive reserve speak to whom and, therefore, determines whether cognitive control
This term refers to the resistance of certain aspects of cognition to brain mechanisms need to be applied to keep speaking in one language or
damage. For example, neurodegenerative diseases can affect the switch to the other language.
cognitive abilities of different individuals to varying degrees, suggesting Orthography-to‑phonology mappings
that the cognitive abilities of some individuals are more resistant to brain This term refers to the correspondence between letters and sounds. This
damage. Cognitive reserve seems to be related to, among other factors, mapping can be different across languages, leading sometimes to
environmental factors such as lifestyle and education. inconsistencies across them. For example, the letter ‘p’ maps into the
Conflict between the two linguistic systems sound /p/ in English and into /r/ in Russian.
This term refers to the potential competition that the different Phoneme repertoire
representations of the two languages of a bilingual may lead to. Given the This term refers to the set of phonemes spoken in a specific language. It is
parallel activation of the two languages and the consequent activation of also known as the phoneme space.
the two linguistic systems, bilingual speakers need to choose the
representations of the target language while ignoring those of the Phonological regularity distributions
non-intended one. This term refers to the fact that combinations of speech sounds can
convey important information about a language. They may represent
Convergence hypothesis specific combinations of phonemes or other types of speech information.
This hypothesis states that the neural networks involved in language For example, in English, stress is placed on the first syllable in most nouns
acquisition and processing are similar for the first and second languages. (unless they are monosyllables), so syllable stress is a useful cue to
This is not to say that some additional neural resources are not required segment nouns in this language.
when learning and using a second language. Indeed, the second language
may, for example, require the recruitment of certain language control Shift in the phoneme space
neural structures. As a consequence of exposure to a different phonological system, such as
in the case of a second language, bilinguals tend to adapt the phoneme
First language
space of their first language. This results in bilinguals speaking their first
This term refers to the first language (or languages) that an individual
language with an accent. This shift occurs more often when the first
learns. In the case of simultaneous bilinguals, there is more than one first
language is used less frequently than the second language.
language. An equivalent term is ‘native language’. A related concept is
dominant language, which refers to the language that an individual feels Simultaneous bilinguals
more comfortable speaking or is fluent in. It is not uncommon that This term refers to bilinguals who are exposed to both languages from
bilinguals change language dominance following, for example, birth, although occasionally the term is also used to refer to individuals
immigration. Unless specified, in the present review, the term first who acquire their second language within the first 2–3 years of life. Such
language is used to refer to a language that was initially learned and that individuals are also called first-language bilinguals.
has remained the dominant one. Successive bilinguals
Inhibition This term refers to bilinguals who are exposed to a second language after
This term is used to describe a cognitive control mechanism that tunes they have started to acquire their first language. There is no general
out stimuli that are irrelevant to the task at hand. In bilingual consensus on how much of the first language needs to be acquired before
conversations, it prevents the speaker from producing utterances in the second-language learning commences for someone to be a labelled a
undesired language by keeping its lexical nodes under the threshold for successive bilingual. Still, researchers often used the terms ‘early’ and
selection. ’late’ bilingualism to refer to speakers that have acquired the second
language before or after the critical periods associated with language
Lexical retrieval failures
learning (BOX 4), respectively.
This term refers to those speaking instances in which individuals make an
error when producing a word. Such errors may involve saying a The mutual exclusivity heuristic
semantically related word instead of the target one (for example, ‘apple’ This term refers to a strategy that humans (adults and infants) have at
instead of ‘peach’), saying a phonologically related word (for example, their disposal to learn new names of objects. The principle is based on the
‘reach’ instead of ‘peach’) or having difficulties to come up with the assumption that humans think that objects should only have one category
intended word in the absence of any intrusion. This type of difficulty in label. In a basic experimental setup to demonstrate this principle, two
retrieving the intended word, often referred to as a ‘tip-of-the-tongue objects are presented to an individual: one for which the individual knows
state’, seems to be more prevalent in bilinguals than in monolinguals, even the name and the other whose name is unknown. Upon hearing a new
if bilinguals try to retrieve the word in any of their two languages. word, humans tend to assign the new word to the nameless object.
Monitoring Word form
The process of evaluating the need to apply cognitive control mechanisms This term refers to those speech units that listeners can segment and
(for example, inhibition) in response to the current context. In bilingual recognize in the speech signal but for which they may not have a
conversations, monitoring is involved in determining which language to meaning.
and Japanese6. This capacity to differentiate The ability to discriminate between their environment at an early stage and that
between two languages is not restricted to more-similar languages, such as English early exposure to a bilingual environment
humans. Previous studies have reported that and Dutch or Spanish and Italian, develops does not hinder this ability.
cotton-top tamarin monkeys7 as well as a bit later, at around 4–5 months of age, in However, the bilingual experience does
Long-Evans rats8 can discriminate between both monolingual and bilingual infants, if seem to affect the way in which language
Japanese and Dutch. Thus, the initial lan- they have previously been exposed to at least discrimination is achieved. One study 9
guage discrimination capacities of humans one of the languages in question9–11. Thus, involving bilingual and monolingual
may have nothing to do with previous it seems that infants are able to notice that 4–5-month-olds showed that monolin-
exposure to language. there are two different language systems in gual infants orientated faster to a familiar
NATURE REVIEWS | NEUROSCIENCE VOLUME 15 | MAY 2014 | 337

PERSPECTIVES
Box 2 | Methodological considerations in bilingualism research

in fact learn two sets of phonemes18–21. The
time it takes to establish the phoneme reper-
In contrast to learning a first language, the acquisition of a second language can occur in many toire in bilinguals is remarkable given that, in
more contexts. Indeed, second-language learners can vary considerably in terms of the age at monolinguals, low-frequency phonemes take
which second-language acquisition occurs, the amount of exposure to that language, the longer to become established than highly fre-
motivation to learn the language, the type of learning experience and the degree of similarity
quent ones22 (note that bilingual studies have
between their two languages. This natural variation affords the opportunity to test relevant issues
about language learning and brain plasticity, such as the presence of critical periods in language
not explored the acquisition of very infre-
acquisition, but it also causes crucial methodological problems, such as finding homogeneous quently occurring phonemes). Indeed, one
study groups of individuals that share equivalent linguistic experiences. In this regard, may have expected that bilingualism would
computational models might be promising tools to facilitate the identification of relevant variables induce a general delay in the acquisition
affecting second-language processing while controlling for potential confounds102,103. of the phoneme repertoire, as presumably
Some individuals seem to be more ‘talented’ than others when learning a new language104,105. The bilinguals receive less exposure in any one
inherent variability in the factors that affect second-language learning, however, makes it difficult of their languages than monolinguals do in
to determine the origin of the inter-individual differences in second-language proficiency. This their one language.
issue is particularly relevant for neuroimaging studies that typically explore relatively small The second major milestone in language
samples and thus can be more affected by heterogeneous sampling106,107. Moreover, language
development for which substantial data exist
studies involving individuals with neurological conditions (usually brain injury or
neurodegenerative conditions) are especially affected by this variability, as premorbid language
for monolinguals and bilinguals concerns
performance is often unavailable in these cases. word learning. Word learning is a complex
Furthermore, monolinguals and bilinguals often differ in terms of fundamental variables such as process. In a highly simplified way, it can be
socioeconomic status and/or emigration. For example, in the United States, bilingualism is often said that it consists of assigning a concept
associated with low socioeconomic status, as reflected by the fact that legislation referring to to a word form (BOX 1). As described below,
bilingual education is included in a Federal programme for disadvantaged students101. The a priori, the bilingual experience may affect
correlation between emigration and bilingualism may, however, also be influenced by the ability to determine word forms and
self-selection, as individuals who emigrate may be more capable or ambitious than others who do assign concepts to them.
not (to make the comparison even more complex, there are differences between emigrants who The identification of words forms in
move to improve economic opportunities and refugees who are forced to emigrate108).
speech is heavily dependent on the computa-
tion of phonological regularity distributions
language than to an unfamiliar one, whereas to perceive and track relevant information in in the absence of the clear word boundaries
bilingual infants showed the opposite pat- two different systems (FIG. 1). in spoken (but not written) language (BOX 1).
tern (FIG. 1). At present, we do not have In summary, these results reveal that One of the regularities that monolingual and
a clear understanding of the underlying monolingual and bilingual infants show bilingual infants have to extract is the com-
mechanisms that lead to such differences. similar developmental trajectories in terms binations of phonemes that signal word end-
The existence of differences between the of language discrimination; however, the ings. For example, infants learning English
mechanisms that underlie language discrimi- bilingual input seems to tune some of the will eventually learn that ‘tr’ does not occur
nation in bilinguals and monolinguals is sup- mechanisms behind this ability. at the word end, although it can occur at
ported by other observations. Information the word onset. Similarly, ‘rt’ can be a word
about the properties of speech is conveyed Establishment of the phoneme repertoire offset but not a word onset. Thus, ‘purt’ but
not only by sounds but also by articulatory and word learning. The fact that bilingual- not ‘putr’ would conform to the pattern of an
gestures12,13. Interestingly, two studies have ism does not hinder the ability to discrimi- English word (even though ‘purt’ is not in fact
reported that bilinguals and monolinguals nate between languages does not necessarily an English word). In the second half of the
differ in their capacity to discriminate imply that a bilingual experience does not first year of life, monolingual infants start
between different languages when watch- affect language learning. Investigation of showing sensitivity to these kinds of proper-
ing silent videos of people speaking these language learning in this context began ties in the words of their native language23.
languages14,15. One of these studies showed relatively recently and has primarily focused By 9 months of age, monolingual infants
that French- or English-speaking monolin- on two crucial issues: the acquisition of the discriminate sequences of sounds that occur
gual infants and French–English-speaking phonetic system and early word learning. in their native language from sequences that
bilingual infants (aged 4 and 6 months old) One of the best-described early stages of do not. One study compared the abilities of
were able to discriminate people speaking monolingual development is the establish- 10‑month-old monolinguals and bilinguals
French from those speaking English in silent ment of the phoneme repertoire3,16. Most to differentiate possible from impossible
videos. Crucially, however, this ability was phonemes become established during the word endings24. Monolinguals and bilinguals
only retained at 8 months of age by the bilin- second half of the first year of life, and on showed an equivalent capacity to differenti-
gual infants14. In the other study, the same the approach to this milestone, infants show ate possible from impossible word endings
silent videos were presented to 8‑month-old decreasing sensitivity to speech sounds that in their shared language provided that it was
Spanish- or Catalan-speaking monolinguals are not present in their environment and the dominant language in the bilinguals’
and Spanish–Catalan bilingual infants who an increasing sensitivity to speech sounds environment. If bilinguals were tested in
were never previously exposed to French or that are associated with the language (or their non-dominant language, then mono-
English. Again, only bilinguals were able to languages) that they are exposed to17. The linguals showed greater discrimination of
discriminate between the two languages15. available evidence reveals that there are no possible word endings than bilinguals. These
Together, these observations suggest that major differences in the time required by data suggest that the identification of words
bilingual infants show a specific adaptation bilinguals and monolinguals to establish their in speech may be affected in bilinguals, but
in the attentional system that enables them phoneme repertoires, even though bilinguals only in individuals for whom there is a clear

PERSPECTIVES
Box 3 | The neural representation of two languages

word meanings in a given language may differ
slightly between monolinguals and bilin-
How bilinguals represent and manage their two linguistic systems is a core issue in bilingualism. guals. At a lexico-syntactic level, bilinguals
The received wisdom is that certain linguistic representations and processes seem to be shared retrieve and utter words slower than monolin-
across languages and that the two languages are active in parallel in most contexts109–114. guals and tend to experience more tip‑of‑the-
Indeed, similar brain structures are involved when bilinguals use either of their two
tongue states than do monolinguals33–35
languages115. Consistent with the ‘convergence hypothesis’ (REF. 116) (BOX 1), the degree of
(BOX 1). Also, bilinguals produce fewer words
neural overlap between the two languages primarily depends on second-language proficiency
and to some extent on the age of onset of second language acquisition. Also, it seems that the of a given semantic category in fluency tasks
linguistic principles governing the organization of the two languages are the same117,118. than do monolinguals36. Together, these
However, there is also evidence suggesting that some of the brain areas that are involved in results show that, compared with mono-
language control are differentially recruited during first-language versus second-language use, linguals, bilinguals exhibit reduced speech
and this is often attributed to the greater effort associated with second-language processing fluency. Furthermore, the frequency of syn-
rather than to differences in the actual representation of the two languages53,119,120. tactic constructions in the second language
Taken together, these results indicate that, in bilinguals, the neural circuits that ‘house’ their two affects bilinguals’ syntactic choices in their
languages are similar. Indeed, cortical regions in the left perisylvian areas, including specific first language37. At the phonological level,
frontal, temporal and parietal regions, together with some subcortical structures (such as the basal
bilinguals’ phoneme boundaries differ from
ganglia) seem to be functionally specialized in the processing of language computations, both for
the first and second languages.
those of monolinguals’. Thus, use of a second
language shifts the phoneme space (as has
been observed in individuals who have been
difference between the exposure times to the vocabularies than monolinguals know in living in a foreign country for a long period
two languages and only then in their more their one vocabulary 29,30). of time) (BOX 1). However, this does not mean
infrequently used language. To conclude, we have argued that, on that bilinguals cannot have two phonological
As mentioned above, word learning also the basis of currently available data, mono repertoires, one for each language38.
involves linking word forms to concepts. lingual and bilingual infants are comparable Three main origins of these effects of
This process may be guided by the so‑called in their capacities to discriminate between bilingualism have been proposed. First,
mutual exclusivity heuristic, according to languages, to learn phoneme repertoires some of these effects might arise because
which individuals hypothesize that new and to learn words. Certainly, the available individuals who become bilingual start
words correspond to new concepts25 (BOX 1). evidence suggests that bilingualism does not using their first language less frequently
In support of this principle, when presented seem to compromise language acquisition than monolinguals use their language.
with a known and an unknown object (the markedly. Nevertheless, some studies show Indeed, studies suggest that the extent to
concept in this case) while listening to an that the bilingual input induces some specific which processing of the first language is
unknown word, monolingual toddlers tend adaptations in the mechanisms underlying affected might correlate with the frequency
to look at the unknown object for a longer such achievements. The precise nature of of second-language usage34,39,40.
period of time than the known object 25. The these adaptations and whether they reflect Second, the effects of bilingualism on
mutual exclusivity heuristic is a useful learn- enhanced language processing in bilinguals first-language processing may arise because
ing strategy given that objects tend to be compared with monolinguals are yet to be of the continuous interaction between the
labelled with just one word (synonyms such determined. first-language and second-language systems,
as ‘couch’ and ‘sofa’ are rare). leading to linguistic ‘transfer’ from the second
Interestingly, the outcome in the above Language processing to the first language32. For example, lexical
task is different in bilingual infants: when In this section, we review studies that have and syntactic choices (use of passive or active
presented with a known and an unknown explored how learning and using a second constructions) in the first language might be
object while listening to an unknown word, language may modify or interfere with first- modulated by the lexical and syntactic prop-
bilingual toddlers look at both objects for language processing in adult populations (the erties of the second language37. Interestingly,
similar periods of time, suggesting that ‘bilingual effect’). in extreme situations, linguistic transfer along
they are not using the mutual exclusivity with a reduced use of the first language can
heuristic26–28. This is probably explained Behavioural consequences of the bilin- lead to first-language attrition41, such as in the
by the fact that each object is likely to be gual effect on first-language processing. case with international adoptions, in which
linked to two labels — one in each lan- Bilingualism can have consequences for the first language is often no longer (or very
guage — rather than to just one, and thus performance in the first language at various infrequently) used42.
the mutual exclusivity heuristic may not levels of linguistic processing (see REF. 31 for Last, the bilingual effect on first-language
be helpful in a multilingual setting. How a review). These consequences have largely processing might come about because of
bilinguals compensate for the lack of utility been identified in studies involving succes- the need to control and monitor the two
of this principle during word learning and sive bilinguals, in which the performances languages, especially in speech production
whether early vocabulary development is of adult bilinguals in their first language in tasks43,44. Although language processing
compromised in some way by its absence a wide range of tasks have been compared generally involves control and monitoring
remain unclear. However, whatever strat- with those of adult monolinguals. processes (BOX 1), bilingual language process-
egy is implemented, it does not seem to At a semantic level, the mapping of meaning is more taxing on these processes because
compromise word learning, as monolingual ings to lexical items in one language may be bilinguals need to ensure that the right lan-
and bilingual toddlers know a comparable influenced by the way in which correspond- guage is used in each communicative context.
number of words (although, consequently, ing lexical items in a bilingual’s other lan- Furthermore, these control processes are
bilinguals know fewer words in one of their guage are mapped into semantics32. That is, especially important as bilinguals activate

PERSPECTIVES
(BOX 1) of a second language begin to show

Box 4 | Critical periods in language learning
increased activation of the left ventral pre-
Critical (or sensitive) periods refer to periods of time in which brain structures are especially frontal cortex during reading in their first lan-
sensitive to a specific environmental input, meaning that outside these periods, the magnitude of guage. This increase has been interpreted to
the input needed to elicit changes in the brain dramatically increases121. The existence of critical mean that bilinguals have higher lexical and
periods in language learning — in particular, in second-language acquisition — has been (and still
non-lexical demands during reading in their
is) a debated topic122–125. It is popularly assumed that native-like acquisition of a second language
can be achieved if children are exposed to this second language before puberty (another common
first language than monolinguals51. However,
version of this assumption places the limit at before 7 years of age)126. However, statements of this the conclusions of these studies are limited
sort are an oversimplification, as they do not consider the different aspects of language learning, by the fact that they involved a bilingual
which may have different critical periods. Language requires various types of knowledge and experimental setting in which participants are
computations (for example, auditory perception and abstract rule learning), which are supported presented with stimuli from two languages.
by different brain structures, with different maturational timings127–129. Of note, the maturation of This setting may engage certain processes (for
structures involved in auditory perception occurs within the first few months of life128,130, whereas example, identifying the language in which
the maturation of the prefrontal structures (involved in planning and rule computations) extends a given item is presented) that are not neces-
well beyond puberty98,128. Thus, the critical period relating to phonological computations that sarily involved when bilinguals process lan-
depend to some extent on auditory processing is quite restricted, but the critical period for
guage in just one of their languages (and that
grammatical rules may extend for much longer. Hence, the issue of critical periods in language
learning needs to be considered in the context of the different linguistic domains.
are clearly not required when monolinguals
The bilingual environment has often been considered to exemplify an enriched environment. As process their only language). Hence, when
a result, one recurring statement that is often encountered when describing bilinguals is that they interpreting the evidence presented below, it
may exhibit delayed closing of sensitive periods compared with monolinguals58. Although this may is important to keep in mind whether each
be the case, the available evidence is far from conclusive, as studies showing delayed closure of particular study has made use of bilingual or
critical periods in bilinguals have compared monolingual and bilingual groups from different monolingual settings.
populations, hence making it difficult to be certain about the origin of such a delay (for example, in Perhaps the most convincing evidence
some cases, socioeconomic status may have been the reason for differences between the study for increased language-processing demands
populations131). Importantly, evidence from animal studies that have linked increased brain in bilinguals compared with monolinguals
plasticity to enriched environments primarily comes from studies analysing the recovery from
comes from a study 52 in which highly profi-
initially extreme deprivation environments. It has been observed132 that enriched environments
(that is, large cages with running wheels and toys) greatly reduce the adverse effects of early
cient early successive bilinguals performed
deprivation and improve visual acuity in adult animals (for a review, see REF.121). Extrapolating the several linguistic tasks only in their first lan-
results from these deprivation studies to the case of differences in input between monolingual and guage. A comparison of brain activity in these
bilinguals remains far-fetched. bilinguals and monolingual controls revealed
that bilinguals exhibit higher activity in five
left-hemisphere language-related brain areas
their two languages in a non-selective way 45–47. Neural consequences of the bilingual effect (dorsal precentral gyrus, pars triangularis,
That is, when planning to name an apple, on first-language processing. Various stud- pars opercularis, superior temporal gyrus and
Spanish–English bilinguals activate both ies have found potential differences in the planum temporale). These differences were
‘manzana’ and ‘apple’ (the words for apple in neural correlates of language processing in evident in tasks involving word retrieval and
Spanish and English, respectively). Moreover, monolingual and bilingual young adults. articulation, such as picture naming and read-
when a bilingual hears a word in one of their As argued below, it is difficult to determine ing aloud but not in receptive language tasks.
languages (for example a Russian–English exactly how these potential neural differ- Interestingly, monolinguals showed increases
bilingual hearing the Russian word ‘marku’), ences relate to the possible explanations of in activity in the same five brain areas when
phonologically related words in both lan- the effects of bilingualism on first-language the language-processing demands in the
guages are activated (such as the English processing. Nevertheless, most studies have naming and reading tasks were increased.
word ‘marker’). Thus, bilinguals almost have interpreted neural differences as indicators Given these results, the authors concluded
two potential lexical candidates (one in each of increased language-processing demands that the main difference between bilingual
language) for each concept that they want to in bilinguals, resulting from either a reduced and monolingual processing relates to the
express, and hence they need to decide con- frequency of language use or a need for increased processing demands faced by
tinuously which one to utter, in accordance greater linguistic control. bilinguals because of the additional need to
with the situation, and to avoid intrusions According to a few studies, some brain control the two languages, the requirement
from the other language (BOX 1). Because of structures show differential activity in mono- to resolve lexical competition and/or the
these additional demands on cognitive pro- linguals and bilinguals, suggesting a bilingual- reduced frequency of articulatory rehearsal.
cesses, one may expect that first-language specific brain activity signature. For example, Consistent with the notion that bilingual-
efficiency becomes affected (for example, by the left inferior frontal cortex shows increased ism taxes control processes, several studies
a decrease in speech rate). activity in simultaneous bilinguals compared have shown a larger involvement of brain
These three explanations for the bilin- with monolinguals during comprehension areas implicated in language control in
gual effect are not mutually exclusive, and tasks48,49, and this differential activity has bilinguals than in monolinguals. For exam-
indeed they may all contribute to some of been suggested to be involved in some sort of ple, Abutalebi and collaborators53–56 have
the observed effects. Having discussed the language separation mechanism in bilinguals convincingly argued that the head of the left
behavioural consequences of bilingualism for (see REF. 50 for evidence of the involvement of caudate and the left anterior cingulate cortex
first-language processing, we now turn to the the left head of caudate in bilingual language are preferentially recruited during bilingual
issue of how bilingualism affects the neural control). Moreover, individuals that learn the language processing in highly proficient early
circuitry involved in such processing. new orthography-to‑phonology mappings bilinguals. Furthermore, these authors suggest

PERSPECTIVES
a fundamental frequencies, suggesting that

Focal point 1,600
bilinguals have more efficient and flexible
Orientation times (ms)

Speaker
auditory processing than monolinguals58.
1,400 An exciting recent discovery in this con-
text is the fact that bilingualism also seems
1,200 to affect the structure of certain brain areas.
Monolingual For example, early and late highly proficient
Bilingual bilinguals, on average, show increased grey
1,000
Maternal English matter in areas involved in verbal fluency
Language of the sentence tasks (left inferior parietal structures59),
2.0
articulatory and phonological processes
b (left putamen60) and auditory processing
1.5 (Heschl’s gyrus61). Furthermore, changes in
white matter tracts have also been reported
Looking times (s)

1.0 to be associated with bilingualism62. A study
of older highly proficient successive bilingual
0.5
adults (70‑year-olds) reported higher white
matter integrity in the corpus callosum in
0.0 (Spanish or Catalan)
bilinguals than in monolinguals63. Some of
Monolingual
(French–English)
Bilingual
(Spanish–Catalan
Bilingual
–0.5 these structural changes are also sensitive to
the proficiency level in the second language,
(French )
–1.0 Monolingual further suggesting that they are indeed
related to the use of a second language rather
than to other potentially uncontrolled vari-
Subjects
ables59. Thus, although it is difficult to give
a complete and coherent picture of the rela-
Figure 1 | Two experiments comparing monolingual and bilingual Nature infants’Reviews
capacities to discrimi-
| Neuroscience tionships between some of these structural
nate between languages. a | In the first study, Spanish or Catalan monolingual and Spanish–Catalan changes and their functional roles, it seems
bilingual 4–5‑month-old infants were presented with English sentences (an unknown language to the
that the learning and continuous use of two
infants) or with sentences of their maternal language (either Spanish or Catalan)9. These sentences
from each language were randomly presented from one of two loudspeakers that were hidden behind languages have pervasive effects on the func-
pictures of women that were arranged on either side of a central screen (two colourful, dynamic tional and structural properties of various
images were successively displayed on the screen at the beginning of each trial (that is, before a sen‑ cortical and subcortical structures.
tence was played) to attract infants’ attention). Previous research with monolingual infants150 had The current evidence is sufficiently
established that infants orientate their gaze faster to the familiar than to the unfamiliar language. consistent to suggest that bilingualism does
Indeed, monolingual Catalan and monolingual Spanish infants displayed the expected pattern (in indeed have behavioural and neural func-
blue). However, Spanish–Catalan bilingual infants showed the opposite pattern: they orientated faster tional consequences for language processing
to the unknown language than to the maternal one (in red). b | In other studies, infants were familiar‑ even in a bilingual’s first and dominant lan-
ized with silent video clips of individuals speaking either French or English. Infants first saw silent video guage. This is not to say, however, that such
clips of three different French–English bilingual speakers speaking either in French or in English and
an experience leads to fundamental differ-
their attention to the images was measured (looking times). Once their attention declined (habituation
criterion), half of the infants saw new sentences in the same language and half of the infants saw new ences in the way that the first language is pro-
sentences in the other language. At the test phase, monolingual and bilingual infants before the age cessed, unless extensive exposure to a second
of 8 months looked longer when presented with video clips from a different language from the one in language causes first-language attrition.
the familiarization phase, indicating that all infants were able to discriminate between the languages.
However, at 8 months of age, only bilingual infants seemed to be able to discriminate between the Beyond language
silent video clips14,15. Furthermore, previous experience with the languages in the silent video clips did As we have discussed, bilingualism affects
not seem to have any marked influence on the capacity of bilingual 8‑month-olds to discriminate the brain activity related to language pro-
between them. The capacity to discriminate French from English visually was equivalent for infants cessing, probably as a result of an increase in
exposed to French and English and for infants exposed to Spanish and Catalan. The bar chart shows language-processing demands. Given that
the increment (or decrement) in looking times between the last trials of the habituation phase and the
exchanging linguistic information is one of
trials in the test phase (when a language change was introduced) for monolingual and bilingual infants
aged 8 months old. As can be seen, only the bilingual infants markedly increased their looking times. the most frequent cognitive activities that
The graph in part a is reproduced, with permission, from REF. 9 © (1997) Elsevier. The chart in part b is humans perform, the question arises as to
reproduced, with permission, for REF. 15 © (2012) SAGE Publications. whether bilingualism affects other cognitive
processes. Research on this issue is perhaps
the topic in the field that is receiving the
that both of these structures are involved in that the increase in processing demands most attention from both the scientific
keeping the two languages apart during lan- associated with bilingualism can also lead to community and the general public at the
guage processing, at least in contexts in which some processing benefits. For example, early moment. In this section, we first review the
both languages are engaged (see REF. 57 for highly proficient bilinguals show increased current evidence regarding the behavioural
similar evidence with bimodal bilinguals — subcortical representation of linguistic consequences of bilingualism in relation to
individuals who can sign and speak the same sounds, as revealed by a larger electrical the efficiency with which executive control
language). However, it must be mentioned brain response in the range of the sounds’ processes work, and then we turn to the

PERSPECTIVES
issue of how bilingualism affects the neural prevent interference from the language that executive control processes. These observa-
circuitry that sustains the executive control is not in use, to update working memory tions nicely complement the behavioural
system across the lifespan. continuously and/or to attend to the relevant effects of bilingualism on the executive con-
linguistic features of each language (for exam- trol system reviewed above (see also REF. 31).
Behavioural consequences of the bilingual ple, the different phonological repertoires) The effects of long-life bilingualism on
effect on executive control processes. The mul- when learning the two languages. However, neural circuitry have been shown to promote
tifactorial executive control system involves the link between the processes engaged cognitive reserve in elderly people66. Elderly
processes such as inhibition (BOX 1), flexible during bilingual language production and bilinguals outperform elderly monolinguals
switching between tasks, working memory comprehension and their potential effects on on executive control tasks66,85, despite the fact
and monitoring 64,65, which may be assigned each of these executive components is poorly that bilinguals recruit certain brain areas,
to any given behavioural task to facilitate its understood79,80. such as the left lateral frontal cortex and cin-
completion. It has been hypothesized that Further complicating this picture is the gulate cortex, to a lesser extent than mono-
these domain-general executive control observation that the cognitive effects of linguals86. Also, bilingualism promotes the
mechanisms are recruited in a more taxing bilingualism are already present in infants. maintenance of white matter integrity of the
manner during bilingual than during mono- Simultaneous bilingual infants are able to corpus callosum in elderly people63, a finding
lingual language processing. Hence, con- switch their attention more efficiently than that has further increased our understanding
tinuous recruitment of these mechanisms monolingual infants in non-linguistic tasks of the basis of cognitive reserve (BOXES 1,5).
during bilingual language processing may at the age of 7 months74, and at 18 months In this context, a striking observation
affect the development and efficiency of the of age, they appear to have more-developed that has deservedly captured media atten-
multifactorial executive control system43,55,66. memory generalization processes81. These tion is that bilingualism seems to delay
Some behavioural evidence supports this observations suggest that an explanation the behavioural symptoms associated
hypothesis: bilinguals experience less inter- of the bilingual effect only in terms of the with neurodegenerative disorders such as
ference in conflict resolution tasks than do need to control their two languages during Alzheimer’s disease87–89. The estimated age
monolinguals66–70 (see REF. 31 for an excellent speech production is not tenable anymore, of onset of the disease and the age of the
review of the scope of the bilingual effect given that these infants do not yet engage in first medical appointment related to cogni-
on cognitive control) and, in some contexts, speech production. tive symptoms associated with dementia are
bilinguals seem to be more flexible when The reviewed evidence strongly sug- about 4–5 years later in proficient bilinguals
switching between non-linguistic tasks71,72. gests that bilingualism has behavioural than in monolinguals. This is not to say that
These effects of bilingualism on the execu- consequences for the mechanisms involved bilingualism protects against the develop-
tive control system have been observed in a in executive control processes. Given these ment of neurodegenerative diseases. Rather,
wide range of tasks with little or no linguis- behavioural observations, it is important the symptoms associated with such diseases
tic content, such as Stroop-like tasks66–68,73. to understand the way in which bilingual- may be delayed in bilinguals because of the
Furthermore, these effects seem to be present ism alters the neural circuitry that sustains presence of greater cognitive reserve caused
not only in simultaneous bilinguals but also executive control processes. by the bilingual experience. However, caution
in successive ones, and across the lifespan; needs to be exercised when trying to general-
that is, from infancy to elderhood66,74. The effects of bilingualism on executive con- ize these latter sets of results, as other studies
Nevertheless, there are certain difficulties trol circuits. The study of how bilingualism have either failed to find this protective effect
when interpreting the results outlined above. affects the neural basis of executive control of bilingualism or have identified a weakly
First, serious concerns have been raised about processes has only recently commenced. protective effect90–93.
the robustness and reliability of the reported Nevertheless, these early studies indicate that In summary, there seems to be sufficient
cognitive effects of bilingualism — especially early bilingualism not only alters the func- experimental evidence supporting the notion
in young adults — and in particular about tional involvement of certain brain areas in that bilingualism has an impact on cognition
which of the different control processes the performance of executive control tasks82–84 beyond language processing, especially on
engaged by bilingual language processing but also induces experience-related changes those processes involved in executive control
actually generates these advantages75–78. in brain structure63,83. For example, when per- and their corresponding brain structures.
Second, our current knowledge about the forming non-linguistic switching tasks, early However, why, how and to what extent bilin-
nature of the different components of the bilinguals recruit larger proportions of the left gualism affects these cognitive processes
executive control system and their interac- hemisphere brain areas related to language and the corresponding brain structures is far
tions with each other is rather limited, mak- control, such as the left striatum and the left from being fully understood.
ing it difficult to relate them to the processes inferior frontal lobe, than do monolinguals82.
involved in bilingual language control. Often, Moreover, early bilinguals seem to recruit Conclusions
our understating of the crosstalk between fewer brain resources in conflict monitor- In the first section of this article, we have
the two systems seems to depend on the use ing tasks than monolinguals, as revealed by argued that currently available evidence
of relatively underspecified terms such as a reduction in brain activity in the anterior suggests that bilingualism does not seem to
‘inhibition’ and ‘monitoring’. Third, it is not cingulate cortex 83. Indeed, the anterior cin- compromise language acquisition, although
immediately obvious which (and how many) gulate cortex seems to be specially tuned by the bilingual input seems to induce specific
aspects of bilingualism might enhance execu- bilingualism, given that its grey matter density adaptations in the mechanisms that underlie
tive control processes. The bilingual effect on (volume) is greater in early bilinguals than in this process. In subsequent sections, we have
the executive control system may come about monolinguals83. Thus, neuroimaging studies described how becoming bilingual affects
because of the need to decide which language convincingly show that bilingualism does first-language processing and executive
to use for each particular interlocutor, to have effects on brain structures involved in control processes in adulthood. Across the

PERSPECTIVES
Box 5 | Brain damage and bilingualism

Importantly, other than a handful of
studies investigating international adop-
Since the first theories of language impairment in bilingual speakers were proposed in the tions, few studies have explored the neural
nineteenth century, the issue of how brain damage affects the two languages of bilingual speakers changes associated with switched language
has attracted the attention of neuropsychologists133,134. This research has mostly focused on the dominance. A closer inspection of this issue
differential effects that brain damage may have on the two languages of a bilingual individual.
could facilitate the development of a better
Indeed, several patterns of impairment of the two languages have been described in such
individuals (see REF. 135 for a review). Cases such as these have often been used as the basis of
understanding of how brain plasticity changes
theories about the cortical representation of two languages and the control mechanisms that across the lifespan. Moreover, a more-com-
enable bilingual speakers to activate the intended language at will136,137. Informative as these plete description of how the age at the time
studies have been, there have also been some difficulties in interpreting their results. This is of acquisition of the second language affects
because it has been unclear what the premorbid language characteristics of the bilingual patients first-language processing and executive con-
were and whether their language deficits arose as a consequence of damage to linguistic trol will be fundamental for understanding
knowledge or damage to the language control mechanisms136. the origin of the reported bilingual effects at
Another aspect of the research involving bilingual patients with language deficits after brain the behavioural and neural levels.
injury that has attracted much interest is the type of language treatment that is most appropriate From a neural perspective, an outstand-
for any given patient; that is, should linguistic rehabilitation target the two languages
ing issue that warrants further research is the
simultaneously or, if only one, which one (for example, the first language or the better-preserved
language)? Moreover, is there transfer between the language targeted by the treatment and the
development of brain networks in monolin-
other language? These questions are still debated, and it seems that each particular case may gual and bilingual children. For a long time, it
require different solutions138. Beyond the clear clinical implications that this research has, it can has been assumed that the complex language
also help us to understand the relationship between the cortical representation of two languages network that supports language processing in
and how the brain recovers function after injury139,140. adulthood is the outcome of a temporally pro-
To complement the studies presented above, researchers have started to explore how tracted interaction between brain maturation
neurodegenerative diseases affect linguistic performance in bilinguals. In contrast to linguistic and language exposure. The normal devel-
performance in stroke, linguistic performance in neurodegenerative diseases can be tracked opmental course of language networks has
alongside the pattern of progressive neuronal loss. Moreover, studies of individuals with such been, and indeed largely still is, considered to
diseases provide greater knowledge of the individuals’ premorbid language capacities, enabling
involve increasing functional left-lateraliza-
better controlled analyses of data. At present, research has mostly focused on how Alzheimer’s
disease (AD) and Parkinson’s disease (PD) affect language performance in bilinguals. Interestingly,
tion of and increasing involvement of frontal
given the different neural structures affected by these diseases, these studies can inform us about structures in these networks. However, recent
different aspects of bilingual language processing136,137,141–144. As AD seems to affect neural findings indicate that such networks (both in
structures involved in episodic and semantic memory (at least in its early stages), the study of terms of left-specialization and in the involve-
bilingual individuals with AD can help us to better understand to what extent the representations ment of their frontal structures) may already
of the two languages share a common neural substrate and how robust such representations be functional at birth98, even in premature
are145,146. By contrast, understanding how PD affects linguistic performance in two languages in infants99. We are just starting to understand
bilinguals can inform models of language control. This is because PD primarily affects the how brain networks develop in monolingual
subcortical areas, such as the basal ganglia and their connections with the prefrontal structures infants. This information will be crucial to
(frontostriatal network147,148) that are supposed to be involved in bilingual language control149.
better understand the origin of the neural
However, caution needs to be exercised when drawing generalizations from this research to
individual cases, as these diseases are often rather heterogeneous in terms of the brain structures
effects that bilingualism induces in adults.
that are affected by the disease process and in the pattern of cognitive deficits. Finally, it is important to keep in mind that
advancing our knowledge of how bilingual-
ism sculpts the brain is a socially relevant
three sections, we have proposed that the demands, which starts in infancy and con- issue. Such knowledge will help to debunk
main differences between monolinguals and tinues throughout life, possibly enhancing some misconceptions and ‘neuromyths’ asso-
bilinguals in terms of language acquisition cognitive reserve in the elderly. ciated with bilingualism, such as the belief
and processing are rooted in two factors. However, these conclusions must be inter- that infants exposed simultaneously to two
First, bilinguals receive less exposure to and preted with some caution when designing languages suffer incomplete language acquisi-
make less use of each of their languages than linguistic educational policies and offering tion or that bilinguals are cleverer than mono-
monolinguals do in their only language. parental advice. This is because a comparable linguals100. This is fundamental given the
Second, bilinguals need to monitor their level of competence between bilinguals and controversial nature of the subject, which is
language systems in a more demanding way monolinguals may only be possible if the frequently discussed in the context of socially
than monolinguals, requiring the involve- linguistic input in any language (and par- sensitive issues such as immigration, civil
ment of cognitive control structures. These ticularly the first one) is frequent, varied and rights or education101.
two features increase the processing demands socially useful94. If these conditions are not
Albert Costa and Núria Sebastián-Gallés are at the
during bilingual language acquisition and met, one finds situations of switched language
Center for Brain and Cognition, Department of
processing. Thus, although the neural net- dominance in which the second language of a Technology, Pompeu Fabra University, 08018
works involved in first-language process- bilingual becomes the dominant language95,96. Barcelona, Spain.
ing seem to be fundamentally the same for This is frequently the case in minority and Albert Costa is also at the ICREA (Institució Catalana
monolinguals and bilinguals, the latter group immigrant populations97. A more balanced de Recerca I Estudis Avançats), Passeig Lluís
faces higher processing demands that lead use of the two (or more) languages by a bilin- Companys, 23; 08010 Barcelona, Spain.
to an increase in brain activity. Furthermore, gual individual should warrant full develop- Correspondence to A.C.
a boost in executive control abilities results ment of the first language and possibly of the e-mail: [email protected]
from coping with this increase in processing collateral advantages in cognitive processes80. doi:10.1038/nrn3709

PERSPECTIVES
1. Grosjean, F. What bilingualism is NOT. Multilingual 28. Houston-Price, C., Caloghiris, Z. & Raviglione, E. 53. Abutalebi, J. Neural aspects of second language
Living [online], http://www.multilingualliving. Language experience shapes the development of the representation and language control. Acta Psychol.
com/2011/03/03/what-bilingualism-is-not/ (2010). mutual exclusivity bias. Infancy 15, 125–150 128, 466–478 (2008).
2. Genesee, F. Bilingual first language acquisition: (2010). 54. Abutalebi, J. et al. Language control and lexical
exploring the limits of the language faculty. Annu. Rev. 29. Hoff, E. et al. Dual language exposure and early competition in bilinguals: an event-related fMRI study.
Appl. Linguist. 21, 153–168 (2001). bilingual development. J. Child Lang. 39, 1–27 Cereb. Cortex 18, 1496–1505 (2008).
3. Gervain, J. & Mehler, J. Speech perception and (2012). 55. Abutalebi, J. & Green, D. Bilingual language production:
language acquisition in the first year of life. Annu. Rev. 30. Core, C., Hoff, E., Rumiche, R. & Señor, M. Total and the neurocognition of language representation and
Psychol. 61, 191–218 (2010). conceptual vocabulary in Spanish–English bilinguals control. J. Neurolinguist. 20, 242–275 (2007).
4. Kuhl, P. K. Brain mechanisms in early language from 22 to 30 months: implications for assessment. 56. Garbin, G. et al. Neural bases of language switching in
acquisition. Neuron 67, 713–727 (2010). J. Speech Lang. Hear. Res. 56, 1637–1649 (2013). high and early proficient bilinguals. Brain Lang. 119,
5. Byers-Heinlein, K., Burns, T. C. & Werker, J. F. The 31. Bialystok, E., Craik, F. I. M. & Luk, G. Bilingualism: 129–135 (2011).
roots of bilingualism in newborns. Psychol. Sci. 21, consequences for mind and brain. Trends Cogn. Sci. 57. Zou, L., Ding, G., Abutalebi, J., Shu, H. & Peng, D.
343–348 (2010). 16, 240–250 (2012). Structural plasticity of the left caudate in bimodal
6. Nazzi, T., Bertoncini, J. & Mehler, J. Language 32. Pavlenko, A. & Malt, B. C. Kitchen Russian: cross- bilinguals. Cortex 48, 1197–1206 (2012).
discrimination by newborns: toward an understanding linguistic differences and first-language object naming 58. Krizman, J., Marian, V., Shook, A., Skoe, E. &
of the role of rhythm. J. Exp. Psychol. Hum. Percept. by Russian–English bilinguals. Biling. Lang. Cogn. 14, Kraus, N. Subcortical encoding of sound is enhanced
Perform. 24, 756–766 (1998). 19–45 (2011). in bilinguals and relates to executive function
7. Ramus, F., Hauser, M., Miller, C., Morris, D. & 33. Gollan, T. & Acenas, L. What is a TOT? Cognate and advantages. Proc. Natl Acad. Sci. USA 109,
Mehler, J. Language discrimination by human translation effects on tip‑of‑the-tongue states in 7877–7881 (2012).
newborns and by cotton-top tamarin monkeys. Science Spanish–English and Tagalog–English bilinguals. 59. Mechelli, A. et al. Neurolinguistics: structural plasticity
288, 349–351 (2000). J. Exp. Psychol. Learn. Mem. Cogn. 30, 246–269 in the bilingual brain. Nature 431, 757 (2004).
8. Toro, J., Trobalon, J. & Sebastian-Galles, N. The use of (2004). 60. Abutalebi, J. et al. The role of the left putamen in
prosodic cues in language discrimination tasks by rats. 34. Ivanova, I. & Costa, A. Does bilingualism hamper multilingual language production. Brain Lang. 125,
Animal Cogn. 6, 131–136 (2003). lexical access in speech production? Acta Psychol. 307–315 (2013).
9. Bosch, L. & Sebastian-Galles, N. Native-language 127, 277–288 (2008). 61. Ressel, V. et al. An effect of bilingualism on the auditory
recognition abilities in 4‑month-old infants from 35. Sadat, J., Martin, C. D., Alario, F. X. & Costa, A. cortex. J. Neurosci. 32, 16597–16601 (2012).
monolingual and bilingual environments. Cognition Characterizing the bilingual disadvantage in noun 62. García-Pentón, L., Pérez Fernández, A., Iturria-
65, 33–69 (1997). phrase production. J. Psycholinguist. Res. 41, Medina, Y., Gillon-Dowens, M. & Carreiras, M.
10. Bosch, L. & Sebastian-Galles, N. Evidence of early 159–179 (2012). Anatomical connectivity changes in the bilingual brain.
language discrimination abilities in infants from 36. Gollan, T., Montoya, R. & Werner, G. Semantic and Neuroimage 84, 495–504 (2014).
bilingual environments. Infancy 2, 29–49 (2001). letter fluency in Spanish–English bilinguals. 63. Luk, G., Bialystok, E., Craik, F. I. M. & Grady, C. L.
11. Nazzi, T., Jusczyk, P. W. & Johnson, E. K. Language Neuropsychology 16, 562–576 (2002). Lifelong bilingualism maintains white matter integrity in
discrimination by English-learning 5‑month-olds: 37. Runnqvist, E., Gollan, T. H., Costa, A. & Ferreira, V. S. older adults. J. Neurosci. 31, 16808–16813 (2011).
effects of rhythm and familiarity. J. Mem. Lang. 43, A disadvantage in bilingual sentence production 64. Miyake, A. et al. The unity and diversity of executive
1–19 (2000). modulated by syntactic frequency and similarity across functions and their contributions to complex “frontal
12. Calvert, G. A. et al. Activation of auditory cortex during languages. Cognition 129, 256–263 (2013). lobe” tasks: a latent variable analysis. Cogn. Psychol.
silent lipreading. Science 276, 593–596 (1997). 38. García-Sierra, A., Ramírez-Esparza, N., Silva- 41, 49–100 (2000).
13. van Wassenhove, V., Grant, K. W. & Poeppel, D. Visual Pereyra, J., Siard, J. & Champlin, C. A. Assessing the 65. Shallice, T. & Burgess, P. The domain of supervisory
speech speeds up the neural processing of auditory double phonemic representation in bilingual speakers processes and temporal organization of behaviour.
speech. Proc. Natl Acad. Sci. USA 102, 1181–1186 of Spanish and English: an electrophysiological study. Phil. Trans. R. Soc. Lond. B 351, 1405–1411 (1996).
(2005). Brain Lang. 121, 194–205 (2012). 66. Bialystok, E., Craik, F. I. M., Klein, R. &
14. Weikum, W. M. et al. Visual language discrimination in 39. Gollan, T. H., Montoya, R. I., Cera, C. & Sandoval, T. C. Viswanathan, M. Bilingualism, aging, and cognitive
infancy. Science 316, 1159–1159 (2007). More use almost always means a smaller frequency control: evidence from the Simon task. Psychol. Aging
15. Sebastian-Galles, N., Albareda-Castellot, B., effect: aging, bilingualism, and the weaker links 19, 290–303 (2004).
Weikum, W. M. & Werker, J. F. A bilingual advantage hypothesis. J. Mem. Lang. 58, 787–814 (2008). 67. Costa, A., Hernandez, M. & Sebastian-Galles, N.
in visual language discrimination in infancy. Psychol. 40. Baus, C., Costa, A. & Carreiras, M. On the effects of Bilingualism aids conflict resolution: evidence from the
Sci. 23, 994–999 (2012). second language immersion on first language ANT task. Cognition 106, 59–86 (2008).
16. Kuhl, P. K. Early language acquisition: cracking the production. Acta Psychol. 142, 402–409 (2013). 68. Costa, A., Hernandez, M., Costa-Faidella, J. &
speech code. Nature Rev. Neurosci. 5, 831–843 41. Schmid, M. S. Languages at play: the relevance of L1 Sebastian-Galles, N. On the bilingual advantage in
(2004). attrition to the study of bilingualism. Biling. Lang. conflict processing: now you see it, now you don’t.
17. Kuhl, P. K. et al. Infants show a facilitation effect for Cogn. 13, 1–7 (2010). Cognition 113, 135–149 (2009).
native language phonetic perception between 6 and 42. Pallier, C. et al. Brain imaging of language plasticity in 69. Bialystok, E. & Martin, M. M. Attention and inhibition
12 months. Dev. Sci. 9, F13–F21 (2006). adopted adults: can a second language replace the in bilingual children: evidence from the dimensional
18. Albareda-Castellot, B., Pons, F. & Sebastian-Galles, N. first? Cereb. Cortex 13, 155–161 (2003). change card sort task. Dev. Sci. 7, 325–339 (2004).
The acquisition of phonetic categories in bilingual 43. Green, D. W. Mental control of the bilingual lexico- 70. Martin-Rhee, M. M. & Bialystok, E. The development
infants: new data from an anticipatory eye movement semantic system. Biling. Lang. Cogn. 1, 67–81 of two types of inhibitory control in monolingual and
paradigm. Dev. Sci. 14, 395–401 (2011). (1998). bilingual children. Biling. Lang. Cogn.11, 81–93
19. Burns, T. C., Yoshida, K. A., Hill, K. & Werker, J. F. The 44. Kroll, J. F., Bobb, S. C., Misra, M. & Guo, T. Language (2008).
development of phonetic representation in bilingual selection in bilingual speech: evidence for inhibitory 71. Prior, A. & MacWhinney, B. A bilingual advantage in
and monolingual infants. Appl. Psycholinguist. 28, processes. Acta Psychol. 128, 416–430 (2008). task switching. Biling. Lang. Cogn.13, 253–262 (2010).
455–474 (2007). 45. Colome, A. Lexical activation in bilinguals’ speech 72. Hernández, M., Martin, C. D., Barceló, F. & Costa, A.
20. Sundara, M., Polka, L. & Molnar, M. Development of production: language-specific or language- Where is the bilingual advantage in task-switching?
coronal stop perception: bilingual infants keep pace independent? J. Mem. Lang. 45, 721–736 (2001). J. Mem. Lang. 69, 257–276 (2013).
with their monolingual peers. Cognition 108, 46. Thierry, G. & Wu, Y. J. Brain potentials reveal 73. Stroop, J. R. Studies of interference in serial verbal
232–242 (2008). unconscious translation during foreign-language reactions. J. Exp. Psychol. 18, 643–662 (1935).
21. Petitto, L. A. et al. The “Perceptual Wedge Hypothesis” comprehension. Proc. Natl Acad. Sci. USA 104, 74. Kovacs, A. M. & Mehler, J. Cognitive gains in 7‑month-
as the basis for bilingual babies’ phonetic processing 12530–12535 (2007). old bilingual infants. Proc. Natl Acad. Sci. USA 106,
advantage: new insights from fNIRS brain imaging. 47. Spivey, M. J. & Marian, V. Cross talk between native 6556–6560 (2009).
Brain Lang. 121, 130–143 (2012). and second languages: partial activation of an 75. Hilchey, M. D. & Klein, R. M. Are there bilingual
22. Anderson, J. L., Morgan, J. L. & White, K. S. irrelevant lexicon. Psychol. Sci. 10, 281–284 (1999). advantages on nonlinguistic interference tasks?
A statistical basis for speech sound discrimination. 48. Kovelman, I., Baker, S. A. & Petitto, L. Bilingual and Implications for the plasticity of executive control
Lang. Speech 46, 155–182 (2003). monolingual brains compared: a functional magnetic processes. Psychon. Bull. Rev. 18, 625–658 (2011).
23. Jusczyk, P. W., Friederici, A. D., Wessels, J., resonance imaging investigation of syntactic 76. Paap, K. R. & Greenberg, Z. I. There is no coherent
Svenkerud, V. Y. & Jusczyk, A. M. Infants’ sensitivity to processing and a possible “neural signature” of evidence for a bilingual advantage in executive
the sound patterns of native language words. J. Mem. bilingualis. J. Cogn. Neurosci. 20, 153–169 (2008). processing. Cogn. Psychol. 66, 232–258 (2013).
Lang. 32, 402–420 (1993). 49. Kovelman, I., Shalinsky, M. H., Berens, M. S. & 77. Kousaie, S. & Phillips, N. A. Conflict monitoring and
24. Sebastian-Galles, N. & Bosch, L. Building phonotactic Petitto, L. Shining new light on the brain’s “bilingual resolution: are two languages better than one?
knowledge in bilinguals: role of early exposure. J. Exp. signature”: a functional near infrared spectroscopy Evidence from reaction time and event-related brain
Psychol. Hum. Percept. Perform. 28, 974–989 investigation of semantic processing. Neuroimage 39, potentials. Brain Res. 1446, 71–90 (2012).
(2002). 1457–1471 (2008). 78. Dunabeitia, J. A. et al. The inhibitory advantage in
25. Markman, E. & Wachtel, G. Children’s use of mutual 50. Crinion, J. et al. Language control in the bilingual bilingual children revisited. Exp. Psychol. http://dx.doi.
exclusivity to constrain the meanings of words. Cogn. brain. Science 312, 1537–1540 (2006). org/10.1027/1618-3169/a000243 (2013).
Psychol. 20, 121–157 (1988). 51. Nosarti, C., Mechelli, A., Green, D. W. & Price, C. J. 79. Calabria, M., Hernandez, M., Branzi, F. M. & Costa, A.
26. Byers-Heinlein, K. & Werker, J. Monolingual, bilingual, The impact of second language learning on semantic Qualitative differences between bilingual language
trilingual: infants’ language experience influences the and nonsemantic first language reading. Cereb. Cortex control and executive control: evidence from task-
development of a word-learning heuristic. Dev. Sci. 12, 20, 315–327 (2010). switching. Front. Psychol. 2, 399 (2012).
815–823 (2009). 52. Parker Jones, O. et al. Where, when and why brain 80. Prior, A. & Gollan, T. H. Good language-switchers are
27. Byers-Heinlein, K. & Werker, J. Lexicon structure and activation differs for bilinguals and monolinguals good task-switchers: evidence from Spanish–English
the disambiguation of novel words: evidence from during picture naming and reading aloud. Cereb. and Mandarin–English bilinguals. J. Int.
bilingual infants. Cognition 128, 407–416 (2013). Cortex 22, 892–902 (2012). Neuropsychol. Soc. 17, 682–691 (2011).

PERSPECTIVES
81. Brito, N. & Barr, R. Influence of bilingualism on 106. Tversky, A. & Kahneman, D. K. Belief in the law of 132. Bartoletti, A., Medini, P., Berardi, N. & Maffei, L.
memory generalization during infancy. Dev. Sci. 15, small numbers. Psychol. Bull. 76, 105–110 (1971). Environmental enrichment prevents effects of dark-
812–816 (2012). 107. Button, K. S. et al. Power failure: why small sample rearing in the rat visual cortex. Nature Neurosci. 7,
82. Garbin, G. et al. Bridging language and attention: size undermines the reliability of neuroscience. Nature 215–216 (2004).
brain basis of the impact of bilingualism on cognitive Rev. Neurosci. 14, 365–376 (2013). 133. Pitres, A. Étude sur l’aphasie chez les polyglottes.
control. Neuroimage 53, 1272–1278 (2010). 108. Chiswick, B. R. Are immigrants favorably self-selected? Rev. Méd. 15, 873–899 (in French) (1895).
83. Abutalebi, J. et al. Bilingualism tunes the anterior Am. Econ. Rev. 89, 181–185 (1999). 134. Pearce, J. M. S. A note on aphasia in bilingual
cingulate cortex for conflict monitoring. Cereb. Cortex 109. Hartsuiker, R. J., Pickering, M. J. & Veltkamp, E. Is patients: Pitres’ and Ribot’s laws. Eur. Neurol. 54,
22, 2076–2086 (2012). syntax separate or shared between languages? Cross- 127–131 (2005).
84. Rodriguez-Pujadas, A. et al. Bilinguals use language- linguistic syntactic priming in Spanish–English 135. Paradis, M. in Handbook of Neuropsychology (ed.
control brain areas more than monolinguals to bilinguals. Psychol. Sci. 15, 409–414 (2004). Berndt, R. S.) 69–91 (Elsevier Science, 2001).
perform non-linguistic switching tasks. PLoS ONE 8, 110. Dijkstra, A. & Van Heuven, W. J. B. in Localist 136. Abutalebi, J., Rosa, P. A. D., Tettamanti, M.,
e73028 (2013). Connectionist Approaches to Human Cognition (eds Green, D. W. & Cappa, S. F. Bilingual aphasia and
85. Kave, G., Eyal, N., Shorek, A. & Cohen-Mansfield, J. Grainger, J. & Jacobs, A. M.) 189–225 (Lawrence language control: a follow‑up fMRI and intrinsic
Multilingualism and cognitive state in the oldest old. Erlbaum Associates, 1998). connectivity study. Brain Lang. 109, 141–156 (2009).
Psychol. Aging 23, 70–78 (2008). 111. Kroll, J. F., Bobb, S. C. & Wodniecka, Z. Language 137. Green, D. W. & Abutalebi, J. Understanding the link
86. Gold, B. T., Kim, C., Johnson, N. F., Kryscio, R. J. & selectivity is the exception, not the rule: arguments between bilingual aphasia and language control.
Smith, C. D. Lifelong bilingualism maintains neural against a fixed locus of language selection in J. Neurolinguist. 21, 558–576 (2008).
efficiency for cognitive control in aging. J. Neurosci. bilingual speech. Biling. Lang. Cogn. 9, 119–135 138. Faroqi-Shah, Y., Frymark, T., Mullen, R. & Wang, B.
33, 387–396 (2013). (2006). Effect of treatment for bilingual individuals with
87. Bialystok, E., Craik, F. I. M. & Freedman, M. 112. Klein, D., Zatorre, R. J., Milner, B., Meyer, E. & aphasia: a systematic review of the evidence.
Bilingualism as a protection against the onset of Evans, A. C. Left putaminal activation when speaking a J. Neurolinguist. 23, 319–341 (2010).
symptoms of dementia. Neuropsychologia 45, second language: evidence from PET. Neuroreport 5, 139. Green, D. W. in Handbook of Bilingualism:
459–464 (2007). 2295–2297 (1994). Psycholinguistic Approaches (eds Kroll, J. F. & de
88. Craik, F. I. M., Bialystok, E. & Freedman, M. Delaying 113. Chee, M. W. Dissociating language and word meaning Groot, A. M. B.) 516–530 (Oxford Univ. Press, 2005).
the onset of Alzheimer disease: bilingualism as a form of in the bilingual brain. Trends Cogn. Sci. 10, 527–529 140. Kiran, S., Grasemann, U., Sandberg, C. &
cognitive reserve. Neurology 75, 1726–1729 (2010). (2006). Miikkulainen, R. A computational account of bilingual
89. Schweizer, T. A., Ware, J., Fischer, C. E., Craik, F. I. M. 114. Chee, M. W., Soon, C. S. & Lee, H. L. Common and aphasia rehabilitation. Biling. Lang. Cogn. 16,
& Bialystok, E. Bilingualism as a contributor to segregated neuronal networks for different languages 325–342 (2013).
cognitive reserve: evidence from brain atrophy in revealed using functional magnetic resonance 141. Costa, A. et al. On the parallel deterioration of lexico-
Alzheimer’s disease. Cortex 48, 991–996 (2012). adaptation. J. Cogn. Neurosci. 15, 85–97 (2003). semantic processes in the bilinguals’ two languages:
90. Chertkow, H. et al. Multilingualism (but not always 115. Perani, D. et al. The bilingual brain. Proficiency and evidence from Alzheimer’s disease. Neuropsychologia
bilingualism) delays the onset of Alzheimer disease: age of acquisition of the second language. Brain 121, 50, 740–753 (2012).
evidence from a bilingual community. Alzheimer Dis. 1841–1852 (1998). 142. Gollan, T. H., Salmon, D. P., Montoya, R. I. &
Assoc. Disord. 24, 118–125 (2010). 116. Green, D. W. in The Interface between Syntax and the da Pena, E. Accessibility of the nondominant language
91. Crane, P. K. et al. Use of spoken and written Japanese Lexicon in Second Language Acquisition (eds van in picture naming: a counterintuitive effect of
did not protect Japanese-American men from Hout, R., Hulk, A., Kuiken, F. & Towell, R.) 197–217 dementia on bilingual language production.
cognitive decline in late life. J. Gerontol. B Psychol. (John Benjamins, 2003). Neuropsychologia 48, 1356–1366 (2010).
Sci. Soc. Sci. 65, 654–666 (2010). 117. Kotz, S. A. A critical review of ERP and fMRI evidence 143. Zanini, S. et al. Greater syntactic impairments in
92. Gollan, T. H., Salmon, D. P., Montoya, R. I. & on L2 syntactic processing. Brain Lang. 109, 68–74 native language in bilingual Parkinsonian patients.
Galasko, D. R. Degree of bilingualism predicts age of (2009). J. Neurol. Neurosurg. Psychiatry 75, 1678–1681
diagnosis of Alzheimer’s disease in low-education but 118. Willms, J. L. et al. Language-invariant verb processing (2004).
not in highly educated Hispanics. Neuropsychologia regions in Spanish–English bilinguals. Neuroimage 144. Zanini, S., Tavano, A. & Fabbro, F. Spontaneous
49, 3826–3830 (2011). 57, 251–261 (2011). language production in bilingual Parkinson’s disease
93. Sanders, A. E., Hall, C. B., Katz, M. J. & Lipton, R. B. 119. Golestani, N. et al. Syntax production in bilinguals. patients: evidence of greater phonological,
Non-native language use and risk of incident dementia Neuropsychologia 44, 1029–1040 (2006). morphological and syntactic impairments in native
in the elderly. J. Alzheimers Dis. 29, 99–108 (2012). 120. Luk, G., Green, D. W., Abutalebi, J. & Grady, C. language. Brain Lang. 113, 84–89 (2010).
94. Schmid, M. S. in Language Attrition. Theoretical Cognitive control for language switching in bilinguals: 145. de Souza, L. C., Lehericy, S., Dubois, B., Stella, F. &
Perspectives (eds Köpke, B., Schmid, M. S., Keijzer, M. a quantitative meta-analysis of functional Sarazin, M. Neuroimaging in dementias. Curr. Opin.
& Dostert, S.) 135–153 (John Benjamins, 2007). neuroimaging studies. Lang. Cogn. Process. 27, Psychiatry 25, 473–479 (2012).
95. Dewaele, J. M. in First Language Attrition: 1479–1488 (2012). 146. Ferreira, L. K., Diniz, B. S., Forlenza, O. V.,
Interdisciplinary Perspectives on Methodological 121. Hensch, T. K. Critical period plasticity in local cortical Busatto, G. F. & Zanetti, M. V. Neurostructural
Issues (eds Schmid, M. S., Köpke, B., Keijzer, M. & circuits. Nature Rev. Neurosci. 6, 877–888 (2005). predictors of Alzheimer’s disease: a meta-analysis of
Weilemar, L.) 81–104 (John Benjamins, 2004). 122. Epstein, S., Flynn, S. & Martohardjono, G. Second VBM studies. Neurobiol. Aging 32, 1733–1741
96. Schmid, M. S. & Dusseldorp, E. Quantitative analyses language acquisition: theoretical and experimental (2011).
in a multivariate study of language attrition: the issues in contemporary research. Behav. Brain Sci. 19, 147. Pavese, N. PET studies in Parkinson’s disease motor
impact of extralinguistic factors. Second Lang. Res. 677–714 (1996). and cognitive dysfunction. Parkinsonism Relat. Disord.
26, 125–160 (2010). 123. Li, P. Lexical organization and competition in first and 18, S96–S99 (2012).
97. Schmid, M. S. Language Attrition (Cambridge Univ. second languages: computational and neural 148. Ray, N. J. & Strafella, A. P. The neurobiology and
Press, 2011). mechanisms. Cogn. Sci. 33, 629–664 (2009). neural circuitry of cognitive changes in Parkinson’s
98. Leroy, F. et al. Early maturation of the linguistic dorsal 124. Werker, J. & Tees, R. Speech perception as a window disease revealed by functional neuroimaging. Mov.
pathway in human infants. J. Neurosci. 31, for understanding plasticity and commitment in Disord. 27, 1484–1492 (2012).
1500–1506 (2011). language systems of the brain. Dev. Psychobiol. 46, 149. Wang, X., Wang, Y., Jiang, T., Wang, Y. & Wu, C. Direct
99. Mahmoudzadeh, M. et al. Syllabic discrimination in 233–251 (2005). evidence of the left caudate’s role in bilingual control:
premature human infants prior to complete formation 125. Juffs, A. Second language acquisition. Wiley an intra-operative electrical stimulation study.
of cortical layers. Proc. Natl Acad. Sci. USA 110, Interdiscip. Rev. Cogn. Sci. 2, 277–286 (2011). Neurocase 19, 462–469 (2012).
4846–4851 (2013). 126. Hu, W. Parents take language class into their own 150. Dehaene-Lambertz, G. & Houston, D. Faster
100. Bhattacharjee, Y. Why bilinguals are smarter. The New hands. The New York Times [online], http://www. orientation latencies toward native language in two-
York Times SR12 (18 Mar 2012). nytimes.com/2006/09/30/nyregion/30play.html month old infants. Lang. Speech 41, 21–43 (1998).
101. Wiese, A. & Garcia, E. The Bilingual Education Act: (2006).
language minority students and US Federal education 127. Hickok, G. & Poeppel, D. The cortical organization of Acknowledgements
policy. Int. J. Bilingual Educ. Biling. 4, 229–248 (2010). speech processing. Nature Rev. Neurosci. 8, 393–402 The authors thank J. Abutalebi, D. Green, M. Burgaleta, P. Li,
102. Li, P. Computational modeling of bilingualism: how can (2007). J. Corey, Y. Gilichinskaya and several members of the Center
models tell us more about the bilingual mind? Biling. 128. Huttenlocher, P. R. & Dabholkar, A. S. Regional for Brain and Cognition at Pompeu Fabra University, Spain,
Lang. Cogn. 16, 241–245 (2013). differences in synaptogenesis in human cerebral for their comments on the manuscript. The authors are sup-
103. Dijkstra, T. & van Heuven, W. J. B. Modeling bilingual cortex. J. Comp. Neurol. 387, 167–178 (1997). ported by grants from the European Community’s Seventh
word recognition: past, present and future. Biling. 129. Huttenlocher, P. R. Dendritic and synaptic Framework Programme (FP7/2007-2013): ERG grant agree-
Lang. Cogn. 5, 219–224 (2002). development in human cerebral cortex: time course ment number 323961; Cooperation grant agreement num-
104. Diaz, B., Baus, C., Escera, C., Costa, A. & Sebastian- and critical periods. J. Dev. Neuropsychol. 16, ber 613465 - AThEME), the Spanish Ministerio de Economía
Galles, N. Brain potentials to native phoneme 347–349 (1999). y Competitividad (PSI2011‑23033; PSI2012‑34071;
discrimination reveal the origin of individual 130. Pujol, J. et al. Myelination of language-related areas in Consolider‑Ingenio2010‑CDS‑2007‑00012) and the Catalan
differences in learning the sounds of a second the developing brain. Neurology 66, 339– 343 Government (SGR 2009–1521). N.S.G. received the prize
language. Proc. Natl Acad. Sci. USA 105, (2006). ‘‘ICREA Acadèmia’’ for excellence in research, funded by the
16083–16088 (2008). 131. Garcia-Sierra, A. et al. Bilingual language learning: an Generalitat de Catalunya.
105. Ventura-Campos, N. et al. Spontaneous brain activity ERP study relating early brain responses to speech,
predicts learning ability of foreign sounds. J. Neurosci. language input, and later word production. J. Phon. Competing interests statement
33, 9295–9305 (2013). 39, 546–557 (2011). The authors declare no competing interests.

How Does The Bilingual Experience Sculpt The Brain?

Uploaded by

Copyright:

Available Formats

How Does The Bilingual Experience Sculpt The Brain?

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

How Does The Bilingual Experience Sculpt The Brain?

Uploaded by

Copyright:

Available Formats

PERSPECTIVES

information (such as the location of articles

336 | MAY 2014 | VOLUME 15 www.nature.com/reviews/neuro

© 2014 Macmillan Publishers Limited. All rights reserved

Box 1 | Key terms and concepts

NATURE REVIEWS | NEUROSCIENCE VOLUME 15 | MAY 2014 | 337

© 2014 Macmillan Publishers Limited. All rights reserved

Box 2 | Methodological considerations in bilingualism research

338 | MAY 2014 | VOLUME 15 www.nature.com/reviews/neuro

© 2014 Macmillan Publishers Limited. All rights reserved

Box 3 | The neural representation of two languages

NATURE REVIEWS | NEUROSCIENCE VOLUME 15 | MAY 2014 | 339

© 2014 Macmillan Publishers Limited. All rights reserved

(BOX 1) of a second language begin to show

340 | MAY 2014 | VOLUME 15 www.nature.com/reviews/neuro

© 2014 Macmillan Publishers Limited. All rights reserved

a fundamental frequencies, suggesting that

Orientation times (ms)

Looking times (s)

NATURE REVIEWS | NEUROSCIENCE VOLUME 15 | MAY 2014 | 341

© 2014 Macmillan Publishers Limited. All rights reserved

342 | MAY 2014 | VOLUME 15 www.nature.com/reviews/neuro

© 2014 Macmillan Publishers Limited. All rights reserved

Box 5 | Brain damage and bilingualism

NATURE REVIEWS | NEUROSCIENCE VOLUME 15 | MAY 2014 | 343

© 2014 Macmillan Publishers Limited. All rights reserved

344 | MAY 2014 | VOLUME 15 www.nature.com/reviews/neuro

© 2014 Macmillan Publishers Limited. All rights reserved

NATURE REVIEWS | NEUROSCIENCE VOLUME 15 | MAY 2014 | 345

© 2014 Macmillan Publishers Limited. All rights reserved

You might also like