The Routledge Handbook of Second Language Acquisition and Neurolinguistics

Page 1

DOI: 10.4324/9781003190912-7

USING STRUCTURAL

NEUROIMAGING TO

INVESTIGATE SECOND

LANGUAGE

Eleonora Rossi, Toms Voits, and Vincent DeLuca

Introduction

Despite the past belief that the brain is static, and that it reaches its peak performance in young

adulthood and then steadily declines as age progresses, early seminal research on neuroplasticity in

animal models (Rosenzweig et al., 1962) demonstrated that the brain is quickly pliable, as a conse-

quence of enriched environmental conditions, different task demands, and different life experiences.

Those key animal models’ findings were based on invasive histological methodologies, involving

slicing and analyzing rats’ brains. Despite the high informativity of those results, this invasive type

of experiments would clearly not be ethical in healthy human subjects. However, with the relatively

recent advent and development of neuroimaging methods such as magnetic resonance imaging (MRI),

scientists in the field have been able to study the neuroplasticity of the human brain non- invasively,

at a relatively large scale, and with increasing replicability across laboratories.

The emergence of neuroimaging methods has enabled collecting neural data non- invasively, and

in- real time, and has revealed that the same neuroplasticity observed in early animal models applies

to the human brain. In the last two decades, research has shown that the human brain changes rapidly

and adapts greatly depending on one’s environment and life experiences. For example, neuroplastic

changes in the brain’s grey matter (primarily composed by neurons’ cell bodies composing the most

superficial part of the brain’s cortex) and white matter (primarily composed by the neurons’ axons)

have been demonstrated to occur as a result of long- term engagement in cognitively demanding tasks,

including motor learning (Bengtsson et al., 2005), visual memory (Maguire et al., 2000), and even

higher- level meditation practices (Hern�ndez et al., 2016). Learning any new cognitively demanding

skill places demands on the cognitive/ neural systems that are implicated by it. In response, the brain

is thought to both form new dendritic spines, the formation of which facilitates new neural pathways,

and prune existing ones to handle the cognitive demands associated with the new skill more optimally

and efficiently (Fuchs & Fl�gge, 2014; see Korenar & Pliatsikas, this volume for a theoretical per-

spective of brain plasticity and second language).

Bilingualism and second language (L2) learning are key examples of such a cognitively demanding

and enriching skill. The data on the effects of L2 learning and bilingualism has been shown to have

consequences for (both) language(s) at the linguistic level (e.g., Leivada et al., 2021), but critically

also for brain structure, brain connectivity, and recruitment patterns (Abutalebi et al., 2012; Li et al.,

2015; Pliatsikas, 2019; Rossi, et al., 2017). Substantial evidence in the field supports the hypothesis

that the observed linguistic and structural effects stem from the underlying language competition

Page 2

Eleonora Rossi, Toms Voits, and Vincent DeLuca

and the cognitive demands to manage two (or more) languages (see Kroll et al., 2013; Rothman

et al., 2019). The study of the structure of the human brain, and especially the investigation of neural

adaptations that may occur in response to L2 learning, and/ or bilingualism more generally, has been

increasingly studied in the last two decades (e.g., DeLuca et al., 2020; Li et al., 2014), especially due

to the availability of MRI and other neuroimaging methods that have catalyzed the study of the struc-

ture and function of the human brain. The description of these methodologies and their applications

to the bilingual brain will be the focus of this chapter.

Bilingualism is still too often assumed to be a constant or a perfectly dichotomous variable at

best. However, recent research in the field is increasingly highlighting how bilingualism exists on

a rather vast continuum of contextual factors that come together to form different individual bilin-

gual profiles. For example, variables such as (a) age- of- acquisition; (b) duration of bilingualism;

course) across an array of domains; (d) size and nature of the speech communities; (e) density of an

individual’s linguistic social networks; (f) expected/ normative choice and prestige of the languages in

the society, and many more factors influence the degree of domain general neurocognitive adaptation.

Even though first steps have been taken to unveil how this multi- level variability modulates struc-

tural and functional brain changes (e.g., Gullifer et al., 2018), this question(s) is still very much open

and will most likely represent the next scientific challenge for the bilingualism scientific community.

The goal of this chapter is to describe (a) the major neuroimaging methods that have been utilized

to date to track structural brain changes that occur in response to L2 learning and bilingualism. We

will then exemplify how data from these methodologies have revealed (b) that the brain changes and

adapts to handle the control and processing demands associated with L2 learning and bilingualism,

and (c) that these adaptations continue to facilitate the handling of these demands in the most efficient

way possible. We will then also highlight the strengths and weaknesses of these methodologies, and

discuss future directions for the research in the field.

Structural Neuroimaging: Critical Concepts and Definitions

The human brain is a highly complex organ. It is beyond the scope of this chapter to thoroughly

describe the human neural central system, but we will provide general concepts that will be important

to understand the basic functioning of the neuroimaging methods that we will describe. For an in-

depth description of the neuroanatomy of the human brain we refer the reader to classic neuroanatomy

textbooks (i.e., Kandel et al., 2000).

The brain is composed of individual nerve cells or neurons. A neuron is considered to be the

basic building block of the brain and characteristically consists of the cell nucleus, multiple smaller

processes called dendrites and a (typically) single longer, more prominent process called the axon.

The dendrites link to other nerve cells via synapses and transmit the information to the cell nucleus.

The cell nucleus fires small impulses called action potentials transmitting them to the effector via

the axon. Unlike dendrites and cell bodies, axons are typically covered in myelin— a fatty tissue that

wraps around the axon forming a sheath, which has the function of facilitating electrical conduction in

axons. Clusters of cell nuclei, which are not covered in a myelin sheath, form what is known as grey

matter, due to its pinkish-grey appearance. Grey matter forms the folded outermost layer of the brain

(the cortex), as well as some subcortical structures deep inside the brain, and consists of cell bodies

and dendrites. Myelinated axon fibers, on the other hand, form what is known as white matter, due to

the whitish appearance of myelin. The white matter consists of bundles of axons that, in turn, form

tracts and is primarily responsible for communication of neural signals between different areas of the

brain. Critically, as we will detail below, grey matter and white matter have different physiological

and structural properties that can be capitalized on by MRI to disentangle different brain structures.

Page 3

Using Structural Neuroimaging to Investigate Second Language

MRI started to be used in the early 1980s to capture three- dimensional structural images of the

brain, and has since become the primary neuroimaging technique to study the structure and the

function of the human brain. MRI is a non- invasive imaging technique that provides measures of

brain structure (e.g., anatomy, measures of white matter integrity), and brain function (i.e., func-

tional MRI or fMRI) with an excellent spatial resolution (~1- 2mm). Even though the vast majority

of the MRI neuroimaging literature on language processing has capitalized on the functional aspects

of MRI (see Kousaie & Klein, this volume, for more details on using fMRI to investigate L2), in this

chapter we will focus on understanding how MRI has been used as a technique in and of itself to

study the relative structural neuroplastic changes in brain structure associated with bilingualism and

L2 learning.

MRI relies on the natural magnetic properties of hydrogen atoms in different types of biological

tissues captured by an induced strong external magnetic field (the MRI scanner, which is essen-

tially a very large magnet), to produce high resolution images of the brain. There is an abundance

of water molecules in the brain, of which hydrogen is the most common chemical. Under normal

circumstances, hydrogen protons spin on their axis in a random fashion. However, in the presence

of a constant strong external magnetic field, such as the MRI scanner, the spin axes line up. The

MRI scanner consists of a main magnet and set of transmitting and receiving coils. With spins of the

hydrogen protons lined up to the external magnetic field, a radio frequency (RF) pulse can be emitted

that causes the spin vectors to deflect and absorb energy. During MR image acquisition, the RF pulses

are switched on and off. In the absence of RF pulse, protons realign with the external magnetic field

and, by doing so, release the RF energy that can be picked up by receiver coils in the MRI scanner.

Critically, as different types of tissues (i.e., bone, cartilage, grey matter, white matter, water, etc.)

have different relaxation times, they can then be identified separately within the MRI to form a three-

dimensional image of the brain. Moreover, different scanning sequences capitalize on two types of

relaxation. T1 (or longitudinal) relaxation captures the return of the excited protons to realignment

with the external magnetic field; T2 (or transverse) relaxation captures the decay of transverse mag-

netization (i.e., when the direction of tissue magnetization is at a 90 degrees angle with respect to

the direction of the magnetic field, and is thus in the transverse plane). Contrast and brightness of

different types of tissue in the acquired MRI image will be determined by T1 and T2 relaxation prop-

erties. The images that are acquired with MRI can then be processed to gather measures of structural

integrity, shape, volume, and density of brain areas and the related structures.

Standard anatomical images (e.g., T1- and/ or T2- weighted images) can provide volumetric

measures of grey matter, white matter, subcortical structures, and fluids, such as cerebrospinal fluid

(Kandel et al., 2000). These measures are then analyzed via sophisticated analyses that enable add-

itional structural assessments such as measures of grey matter density or cortical thickness and

gyrification, which we will describe more in depth below. These relatively new types of methods

are used to quantify changes in brain structures, for example neural adaptations after language

immersion and/ or new language learning. MRI also permits the acquisition of sequences that enable

the tracking of the structure of white matter. As we will describe, diffusion tensor imaging examines

white matter integrity by capitalizing on the anisotropy of water flow (i.e., the ability of a substance

to have differential values when measured in different directions) as an indicator of white matter

integrity.

In sum, MRI is the primary method that allows the examination of the structure of the human

brain in a non- invasive manner. It has great spatial resolution and can be easily coupled with other

neuroimaging methods and with behavioral data obtained outside the scanner, which explains the

wide adoption of this method to study the structural effects of bilingualism and L2 learning in

neuroscientific research. In what follows, we first provide a historical excursus of the methods that

inform the structure and changes that occur in the human brain. Then, we present several current

Page 4

Eleonora Rossi, Toms Voits, and Vincent DeLuca

neurocognitive methods used in bilingualism/ L2 research, and a description of their advantages and

disadvantages. Finally, we present key studies to exemplify each method, and conclude the chapter

by identifying perspectives for future research in the field.

Historical Perspectives

The investigation of the neural structure of the brain, its changes, and its connection with language is

certainly not as recent as one might assume. It dates back millennia, far preceding the development

of modern neuroimaging techniques. Findings of skulls on which surgeries have been performed on,

date back to over 5,000 years ago (Finger, 2001). Similarly, an Egyptian papyrus dating from 1700

BC (Breasted, 1930) contains the description of 48 medical cases written by an Egyptian surgeon

including the earliest known description of a case of aphasia. This description of an aphasic patient,

pre- dates the famous work on aphasia by Paul Broca (1861) by thousands of years!

More recently, starting from the nineteenth century, the neuroanatomy of the human brain has been

investigated mainly through the study of clinical lesions. Abundant anatomo- clinical reports have

provided fundamental data for the structural localization of various cognitive functions, including a

“map” of areas dedicated to language processing (for a review, see Luzzatti & Whitaker, 1996). The

anatomo- clinical method yielded information on what areas of the brain subserve specific linguistic

functions, including a rich literature on early cases of bilingual aphasia that permitted the mapping

and theorizing of how multiple languages are represented in the brain (for detailed information on this

topic, see Scimeca et al., this volume). The early anatomo- pathological models that emerged from

this work, have been primarily based on post- mortem brain analyses, and have since been enriched

by years of psycholinguistic research that has investigated the bases of language processing, leading

to the formulation of foundational psycholinguistic models (e.g., Levelt, 1989). Although behav-

ioral psycholinguistic research thrived, the research on the structure of the brain and its linguistic

underpinnings was at a halt, mainly due to the lack of technology that enabled researching the struc-

ture (and changes thereof) of the human brain in vivo, beyond the analysis of post- mortem brains.

Even though the idea that blood flow was somehow associated with human brain structure

and function was already proposed in 1878 by an Italian physiologist (Mosso, 1881), and fur-

ther formalized by Roy and Sherrington (1890), it was only in the 1970s with the invention and

introduction of x- ray computed tomography and positron emission tomography (PET) that it was

possible to capture an image of the brain in vivo. In 1982, PET was also used to track changes in

blood flow during a behavioral task with a relatively good temporal resolution. However, it was

with the introduction of MRI in 1982, with its ability to track the various structures of the brain

non- invasively, and later create clear images of the human brain “at work” with fMRI (in the 1990s)

that the doors to what we know today as modern cognitive neuroscience were opened. Since the

advent of (f)MRI, and its applications to human neurocognition, the literature in the field of the

neurocognition of language (including bilingualism) and its structural underpinning have exploded.

For a detailed account of the various technological advances of human brain mapping, we refer to

Raichle’s review (2009).

Methods and Paradigms

In this section, we describe the current MRI methodologies that are used to study structural changes

in the human brain. We start from measures to study grey matter adaptation to language learning

experience, and we then continue to methodologies used to capture structural white matter changes.

For each method, we also provide a short section describing benefits and disadvantages. A visual

summary of the described methods is provided in Figure 5.1.

Page 5

Using Structural Neuroimaging to Investigate Second Language

Structural Neuroimaging: Methods for Studying Grey Matter Adaptation

A variety of techniques currently exist that have been used to study patterns of grey matter struc-

ture adaptations to L2 learning and bilingual engagement (Luk et al., 2020). Herein, we focus on

describing the bases of three more commonly used techniques in the field.

Voxel- based morphometry (VBM) is one of the primary techniques used to assess neural grey

matter volume and density. In general terms, VBM compares grey matter concentrations per voxel

across the brain, either between groups, or across subjects (Ashburner & Friston, 2000). In VBM ana-

lysis, subjects’ brains are first segmented into distinct tissue types, then registered to a study- specific

template, and then to standard space. The images are then spatially smoothed to increase across-

subject comparability. Regional signal intensities are then used to calculate grey matter density within

voxels.

Benefits and disadvantages: A primary benefit of VBM is the ability to provide robust measurements

of grey matter volume and/ or density across the whole brain, including the cerebellum and (in prin-

ciple) subcortical structures. This method, however, also carries some limitations. First, the requisite

registration to a standard space means warping of some brain regions. Furthermore, given that some

degree of spatial smoothing is required, this technique may be unable to capture some smaller,

more granular adaptations, for example, shape changes seen within the subcortical structures (e.g.,

Burgaleta et al., 2016; Pliatsikas et al., 2017).

Cortical thickness (CT) is another technique developed for examining regional morphological

patterns of grey matter (Ad- Dab’bagh et al., 2005; Fischl & Dale, 2000). Differently than VBM,

which is specific to measure volume density, CT establishes the extent of grey matter as defined

between the white matter and pial surface (i.e., the surface representing the boundary between grey

matter and cerebrospinal fluid) within the cortex and uses these boundaries to establish the thickness

of the cortex at any given point. Akin to VBM, subject brains are also registered to standard space.

White matter intensity measures are then used to determine the extent of white matter across the

volume, with values being assigned to (a) white matter or (b) other tissue type. A surface is then

Figure 5.1 Summary of the key methods that are used to study brain structural changes in L2 acquisition and

bilingualism.

Page 6

Eleonora Rossi, Toms Voits, and Vincent DeLuca

generated based on this segmentation and refined based on intensity gradients per tissue type. Refined

white matter and pial surfaces are then laid over the T1 image and distance between these is calculated

to give thickness values across the cortex.

Benefits and disadvantages: A primary benefit of CT is that it allows for a granular view of specific

regional fluctuations across the cortex, as it is more sensitive to regional folding patterns within sulci

and gyri. This said, it also carries some limitations. First, like VBM, whole- brain CT analyses typic-

ally require registration to standard space for individual- or group- level comparisons, which entails

some degree of warping. Furthermore, as the name implies, the subcortical structures and cerebellum

cannot be included in this analysis, limiting the scope of examination.

Finally, vertex- based analysis (VBA) was developed specifically for regional shape adaptations

within several of the subcortical regions including the thalamus, caudate nucleus, putamen, hippo-

campus, amygdala, and others (Patenaude et al., 2011). Via a Bayesian framework of shape, subcor-

tical structures are extracted and registered to standard space. Coordinate maps for each subject are

then projected onto the standard template for each structure. The resulting spatial maps from this

projection signify positive and negative displacement values, which are used as a measure of shape

deviation.

Benefits and disadvantages: The primary advantage of VBA is its degree of granularity in exam-

ining structural patterns and changes. This is particularly useful within the subcortical structures

where regional patterns of adaptations may be too subtle to be captured in a volumetric analysis

(like VBM for example). Of course, this technique also carries some limitations. First, like CT and

VBM, it also requires registration to standard space, which entails a degree of warping. Furthermore,

depending on the package used, this technique might only be useful in, or realistically applicable to,

the subcortical structures.

Structural Neuroimaging: Techniques for Studying White Matter Adaptation

MRI can be used to investigate structural changes not only in grey matter, but also in white matter.

Diffusion tensor imaging (DTI) is a MR technique that captures directional water displacements in

white matter tracts (Le Bihan et al., 2001), and is used to investigate changes in white matter structure

an integrity. The basic idea behind DTI is that water tends to flow anisotropically (i.e., directionally

dependent) when it is constrained by a biological “vector,” for example by the myelin sheath, and

isotropically (not directionally dependent) when it is not constrained. For example, cerebrospinal

fluid, which is not constrained, flows relatively isotropically in the ventricles within the brain and in

the spinal cord (Le Bihan et al., 2001).

DTI data is collected using specific MRI recording sequences. The duration of the sequence ranges

from four to ten minutes, depending on the resolution and scanner characteristics. The statistical

model that describes the three- dimensional diffusivity direction of each white matter voxel is called

a tensor. The tensor model in each voxel allows the calculation of levels of diffusivity and anisotropy

that describe the shape of the white matter tract using water displacement as a measurement. DTI data

enables calculating common measures of water diffusion in the white matter tracts, such as: (a) axial

diffusivity, indicating diffusivity parallel to a white matter tract; (b) radial diffusivity, indicating the

diffusivity of water molecules perpendicular to a white matter tract; (c) mean diffusivity, representing

the average diffusivity across all directions; and finally (d) fractional anisotropy, indexing whether

the diffusion is isotropic (when the value is 0, indicating non- directional diffusion) or anisotropic

(when the value is 1, indicating complete directional diffusion). It is important to remember though,

that the interpretation of fractional anisotropy values should be combined with the knowledge of the

anatomical structure of the white matter architecture, given that white matter fibers cross in specific

directions. Degrees of fractional anisotropy can thus be measured across many directions depending

Page 7

Using Structural Neuroimaging to Investigate Second Language

on the direction of the white matter fibers to ascertain the directionality of water flow in each voxel

of the brain (Le Bihan et al., 2001). To date, the standard is to measure mean diffusivity in more than

six directions (measured in degrees). This type of information from each voxel can then be combined

with tractography measures to provide an estimate of longer- range directionalities (e.g., tracts), using

tract- based spatial statistics.

Benefits and disadvantages: A primary benefit of DTI is that it allows for several robust

measurements of white matter microstructure across the brain. Like the other methods described

above, however, this method also carries some limitations to its application. First, due to the way

in which diffusion tensors are calculated, the method is somewhat limited in accuracy in meas-

uring white matter microstructure in areas where fiber tracts cross. Second, it is unable to be used to

measure structural connectivity between brain regions.

Structural Neuroimaging to Study L2 and Bilingualism: Example Studies

In this section, we discuss one key example for each of the above- described methodologies to show

how they have been applied to examine the effects of L2 learning and/ or bilingual experience on

the structure of the human brain. Here, we focus on studies that tested young adults. Note, however,

that the findings might not generalize to the intersection of bilingualism in populations that undergo

developmental brain changes (e.g., in childhood and in older adulthood). See the Further Readings

section below for additional, empirical work for each method.

Voxel- Based Morphometry

M�rtensson and colleagues (M�rtensson et al., 2012) used VBM to examine brain plasticity over

early stages of non- native language learning. In a longitudinal design, participants were tested in

two separate scan sessions, an initial, baseline scan, and an additional scan after three months of an

intensive language acquisition program. The participants were young adults in the Swedish mili-

tary training to be interpreters. A control group was also included in the study that was composed

of university students. The results demonstrated that grey matter volumes in the left superior tem-

poral gyrus, left middle frontal gyrus, and right hippocampus showed significant increases in the

three months post- learning for the interpreters but crucially not for the control group. Hippocampal

volumes also increased significantly more over the three- month period for the interpreters relative

to the controls. Furthermore, within the interpreters, proficiency in language acquisition (measured

by participants’ course performance) correlated positively to grey matter volume in the left superior

temporal gyrus and right hippocampus, whereas increased effort to reach language learning goals

correlated positively with grey matter volume in the left middle frontal gyrus. Altogether, this study

provides evidence that engaging in L2 learning (in this case assumingly at an intense pace), catalyzes

rapid neural grey matter volumetric changes, suggesting a general effect of L2 learning in neural

reshaping.

Cortical Thickness

CT has been utilized as a key measure to study how bilingualism can shape the human brain. Here

we showcase a recent key example. Klein and colleagues (2014) examined the effects of bilingualism

and L2 age of acquisition (AoA) on regional cortical thickness. Three groups of participants were

tested: monolinguals, simultaneous bilinguals, and sequential bilinguals who learned an L2 later in

childhood (8– 13 years). The results revealed that late bilinguals had greater cortical thickness (CT)

in the left inferior frontal gyrus (specifically, pars triangularis and orbitalis) than monolinguals and

Page 8

Eleonora Rossi, Toms Voits, and Vincent DeLuca

early bilinguals, accompanied by a reduction in CT in the right inferior frontal gyrus. Interestingly,

AoA and CT were positively correlated in the left inferior frontal gyrus and negatively correlated in

the right inferior frontal gyrus. That is, the later the AoA, the thicker the left inferior frontal gyrus

and thinner the right inferior frontal gyrus. CT in the left superior parietal lobe was also positively

correlated with AoA. The authors concluded that length of bilingualism has a differential effect on

changes in measures of CT, and that length of length of L2 use modulates CT.

Vertex- Based Analysis

Pliatsikas and colleagues (2017) assessed the effects of L2 immersion on adaptations within the sub-

cortical structures using a vertex- based analysis, which is not easily addressable via cortical thickness

analyses. They tested two experimental populations. Both were L2 speakers of English living in the

UK at the time of testing but differed primarily in their length of residence in the UK. The immersed

bilingual group had a mean length of residence of 7.6 years, whereas the low- immersion group had

a mean length of residence of 3.9 years. Both groups were compared to monolingual controls. Two

effects were examined: (a) group effects (comparing bilinguals to monolinguals); and (b) continuous

effects of language experience, i.e., proficiency, age of acquisition (years), and immersion time (for

the bilingual population only). Comparing bilinguals (with both short and long immersion time) to

the monolinguals, significant shape changes (expansions) were found in the thalamus, globus pal-

lidus, and putamen. Regressions with measures of language experience showed that length of L2

immersion predicted shape change in the globus pallidus bilaterally. Comparing the low- immersion

bilinguals to the monolinguals, shape changes (both expansions and contractions) were found in the

caudate nucleus, but no significant effects for any of the language experiences for this group were

found. The results indicate that prolonged L2 exposure in an immersive context supports increased

measurable efficiency.

Diffusion- Tensor Imaging

Kuhl et al. (2016) examined the effects of L2 immersion on white matter integrity. Two groups of

participants were scanned: young adult Spanish– English bilinguals, and a native- English speaking

control group. All participants were residing in the United States at time of testing. The bilingual

participants started learning English upon moving to the United States. Participants completed a lan-

guage background questionnaire and were scanned with a DTI sequence. Fractional anisotropy and

mean diffusivity values were calculated and compared across participants using the specific dedicated

pipelines in MRI software packages. Generally, higher fractional anisotropy and lower mean diffu-

sivity values were observed for the bilingual group in frontal tracts, while lower values were seen in

posterior tracts. In relation to length of immersion in the US, a negative correlation was found for

all diffusivity values and a positive correlation for fractional anisotropy values, suggesting that the

longer the immersion in the L2 environment, the greater the change in white matter. Furthermore,

mean diffusivity in anterior tracts of the left hemisphere was modulated by increased L2 exposure

(listening), whereas production (speaking) was found to modulate fractional anisotropy values in the

posterior section of the left hemisphere.

Innovations and Future Directions

Even though there is a sizable body of literature, the literature covering neuroanatomical adaptations

to L2 (and bilingualism more generally) is still relatively young, and much remains to be explored.

Indeed, several study designs and methodological innovations could bring novel insights to this

Page 9

Using Structural Neuroimaging to Investigate Second Language

topic, some of which we address here. Herein, we highlight several directions that could be used

to better capture variability in L2/ bilingualism and structural adaptation, to further methodological

developments, and to combine structural imaging with other complementary techniques.

Extending Structural Designs to Better Capture Neural Change

As introduced, bilingualism is increasingly recognized as a multifaceted, rather than monolithic phe-

nomenon. For example, key variables such as age of acquisition, proficiency, and immersion in a L2

environment, the characteristics of that immersive environment, the length of immersion, together

with other emerging variables, including the size and typology of our social network, might shape the

magnitude and type of the observable structural brain changes, and require additional investigations.

One potential strategy to tackle the diversity of these variables and understand how they might shape

neural structural changes would ideally involve large- scale studies across multiple laboratories around

the world. This would allow for more direct comparisons across diverse populations and better cap-

ture the dynamic relationship between specific language use patterns and bilingualism- induced neural

plasticity (Stein et al., 2014).

On a similar note, most work to date comes from group studies looking at a specific type of

bilinguals/ learners at a particular point in time, which, while useful for showing trends across a

population, only provide a single snapshot in time of individuals’ trajectories of adaptation to L2

acquisition and use. One key future direction to better understand these effects would be to study this

trajectory from a long- term (i.e., over multiple years) longitudinal perspective. The key advantage to

using such designs is that they allow for the use of subjects as their own baselines, and thus more dir-

ectly track the nature and degree of structural adaptations to specific aspects of L2, but crucially how

these would dynamically shift over time (see, Pliatsikas, 2020; Korenar & Pliatsikas, this volume).

Extending the concept of longer- term longitudinal research, the applications of such study designs

could be extended to examine the interaction of bilingual experiences with brain development and

cognitive aging. Such cohort studies already exist, though none have data on bilingualism in detail.

Adding a bilingualism component to such research should be relatively easy to do.

Future Directions: Methodological Developments

Methodologies have and will continue to evolve and improve. Given some of the limitations of the

techniques described in this chapter, several methodological advances might prove useful in the fur-

ther study of L2- related neuro- anatomical adaptation. For example, surface- based morphometry

has been shown to provide a more nuanced measure of grey matter volume across the cortex (e.g.,

Del Maschio et al., 2019). Surface- based morphometry is not constrained to the same degree of

smoothing as VBM and thus may be able to provide more fine- grained measures of grey matter adap-

tation (Fischl & Dale, 2000; Pantazis et al., 2010).

Similarly, structural connectivity via tractography or connectometry is another potentially useful

tool for examining adaptations to bilingual experience (for applications of these techniques see Fedeli

et al., 2021; Rahmani et al, 2017), though relatively few studies to date have used it within the field

of neurocognition and bilingualism. Tractography can be used to measure the degree of connectivity

between specific brain regions whereas tract- based spatial statistics provides a metric of white matter

microstructure but cannot be used to measure specific connection between regions, as discussed

above (but see K�hncke et al., 2021 for a multimodal study).

Critically, most studies on L2 and the brain investigate structural adaptations at a macroscopic

level, that is, measuring changes in white matter integrity or grey matter volume/ density at a rank of

individual white matter tracts, cortical areas, or subcortical structures. However, very little is known

Page 10

Eleonora Rossi, Toms Voits, and Vincent DeLuca

about the molecular underpinnings of structural brain change (which is true for general neurosci-

ence). To this end, few recent studies have used various biomarkers as precursors to changes in

structural adaptations in response to L2 experience. Only a handful of studies to date have taken this

approach. Two of these have used magnetic resonance spectroscopy (Pliatsikas et al., 2021; Weekes

et al., 2018), which allows one to measure the concentration of various brain metabolites. Metabolite

levels can serve as indicators for underlying processes supporting dendritic branching and neural

proliferation leading to changes observable on the macroscopic scale. In addition to tapping into

brain metabolites, one study has investigated the link between bilingualism and Alzheimer’s dis-

ease biomarkers in the cerebrospinal fluid in healthy middle- aged individuals (Estanga et al., 2017).

The results demonstrated that bilingual experience moderated the relationship between total- tau

protein concentration and age, such that bilinguals had a more favorable CSF- AD (cerebrospinal

fluid- Alzheimer’s disease) biomarker profile indicating a lower risk to develop Alzheimer’s dis-

ease. Examining metabolites and biomarkers in combination with structural neuroimaging is a

viable future research avenue. As the field advances and structural changes linked to bilingualism

(in its nuance as a set of experiences) become better understood, the underlying mechanisms driving

these adaptations will become subject to empirical investigation. Studying brain structure provides

only one of several aspects related to bilingualism- induced neural plasticity. In sum, by combining

structural, functional, and other methods tapping in the brain, researchers will be able to develop

a more holistic overview and a better understanding of graded brain adaptations in response to L2

experience.

Further Readings

This recent VBM study shows the structural adaptation effects of L2 learning on grey matter volume, revealing

decreased volume in cingulate cortex (ACC) and right inferior frontal gyrus (IFG) after L2 learning for one year.

Importantly, these modulations are modulated by L2 proficiency.

Liu, C., Jiao, L., Timmer, K., & Wang, R. (2021). Structural brain changes with second language learning: A

longitudinal voxel- based morphometry study. Brain and Language, 222, Article 105015. https:// doi.org/

10.1016/ j.bandl.2021.105 015

This recent study compared young bilingual and monolingual adolescents and found that bilinguals had thinner

cortex than monolinguals in several cortical regions. In addition, this study highlights that within bilinguals more

L2 use is reflected in greater CT.

Vaughn, K.A., Nguyen, M.V., Ronderos, J., & Hernandez, A.E. (2021). Cortical thickness in bilingual and mono-

lingual children: Relationships to language use and language skill. NeuroImage, 243, Article 118560. https://

doi.org/ 10.1016/ j.neu roim age.2021.118 560

This recent CT study evaluates the effects of acquiring an L2 both with regards to oral fluency and written abil-

ities and highlights the roles of Age of Acquisition as a modulator of the observed affects.

Tu, L., Niu, M., Pan, X., Hanakawa, T., Liu, X., Lu, Z., Gao, W., Ouyang, D., Zhang, M., Li, S., Wang, J., Jiang,

B., & Huang, R. (2021). Age of acquisition of Mandarin modulates cortical thickness in high- proficient

Cantonese– Mandarin bidialectals. Journal of Psycholinguistic Research, 50, 723– 736. https:// doi.org/

10.1007/ s10 936- 020- 09716- 5

This recent DTI study illustrates the effects of bilingualism on white matter structures across the lifespan and

demonstrates that increased engagement in bilingual language use correlates with a slower decline in white

matter integrity with age.

DeLuca, V., & Voits, T. (2022). Bilingual experience affects white matter integrity across the lifespan.

Neuropsychologia, 169, Article 108191. https:// doi.org/ 10.1016/ j.neuro psyc holo gia.2022.108 191

References

Abutalebi, J., Della Rosa, P.A., Green, D.W., Hernandez, M., Scifo, P., Keim, R., Cappa, S.F., & Costa, A. (2012).

Bilingualism tunes the anterior cingulate cortex for conflict monitoring. Cerebral Cortex, 22(9), 2076– 2086.

https:// doi.org/ 10.1093/ cer cor/ bhr 287

Page 11

Using Structural Neuroimaging to Investigate Second Language

Ad- Dab’bagh, Y., Singh, V., Robbins, S., Lerch, J., Lyttelton, O., Fombonne, E., & Evans, A.C. (2005). Native

space cortical thickness measurement and the absence of correlation to cerebral volume. In K. Zilles (Ed.),

Proceedings of the 11th Annual Meeting of the Organization for Human Brain Mapping.Toronto: NeuroImage.

Ashburner, J., & Friston, K.J. (2000). Voxel- based morphometry— the methods. NeuroImage, 11(6 Pt 1), 805–

821. https:// doi.org/ 10.1006/ nimg.2000.0582

Bengtsson, S.L., Nagy, Z., Skare, S., Forsman, L., Forssberg, H., & Ull�n, F. (2005). Extensive piano practicing

has regionally specific effects on white matter development. Nature Neuroscience, 8(9), 1148– 1150. https://

doi.org/ 10.1038/ nn1 516

Breasted J.H. (1930). The Edwin Smith Papyrus: Published in Facsimile and Hieroglyphic Transliteration with

Translation and Commentary. Chicago, IL: The University of Chicago Press.

Broca, P. (1861). Remarks on the seat of the faculty of articulated language, following an observation of aphemia

(loss of speech). Bulletin de la Soci�t� Anatomique, 6, 330– 357.

Burgaleta, M., Sanju�n, A., Ventura- Campos, N., Sebasti�n- Gall�s, N., & �vila, C. (2016). Bilingualism at the

core of the brain. Structural differences between bilinguals and monolinguals revealed by subcortical shape

analysis. NeuroImage, 125, 437– 445. https:// doi.org/ 10.1016/ j.neu roim age.2015.09.073

Del Maschio, N., Fedeli, D., Sulpizio, S., & Abutalebi, J. (2019). The relationship between bilingual experience

and gyrification in adulthood: A cross- sectional surface- based morphometry study. Brain and Language, 198,

104680. https:// doi.org/ 10.1016/ j.bandl.2019.104 680

DeLuca, V., Segaert, K., Mazaheri, A., & Krott, A. (2020). Understanding bilingual brain function and structure

changes? U bet! A unified bilingual experience trajectory model. Journal of Neurolinguistics, 56, Article

100930. https:// doi.org/ 10.1016/ j.jne urol ing.2020.100 930

DeLuca, V., & Voits, T. (2022). Bilingual experience affects white matter integrity across the lifespan.

Neuropsychologia, 169, Article 108191. https:// doi.org/ 10.1016/ j.neuro psyc holo gia.2022.108 191

Estanga, A., Ecay- Torres, M., Iba�ez, A., Izagirre, A., Villanua, J., Garcia- Sebastian, M., … & Martinez- Lage, P.

(2017). Beneficial effect of bilingualism on Alzheimer’s disease CSF biomarkers and cognition. Neurobiology

of Aging, 50, 144– 151. https:// doi.org/ 10.1016/ j.neu robi olag ing.2016.10.013

Fedeli, D., Del Maschio, N., Sulpizio, S., Rothman, J., & Abutalebi, J. (2021). The bilingual structural

connectome: Dual- language experiential factors modulate distinct cerebral networks. Brain and Language,

220, Article 104978. https:// doi.org/ 10.1016/ j.bandl.2021.104 978

Finger, S. (2001). Origins of neuroscience: A history of explorations into brain function. Oxford University

Press, USA.

Fischl, B., & Dale, A.M. (2000). Measuring the thickness of the human cerebral cortex from magnetic resonance

images. Proceedings of the National Academy of Sciences, 97(20), 11050– 11055. https:// doi.org/ 10.1073/

pnas.200033 797

Fuchs, E., & Fl�gge, G. (2014). Adult neuroplasticity: More than 40 years of research. Neural Plasticity, 2014,

Article 541870. https:// doi.org/ 10.1155/ 2014/ 541 870

Gullifer, J.W., Chai, X.J., Whitford, V., Pivneva, I., Baum, S., Klein, D., & Titone, D. (2018). Bilingual experience

and resting- state brain connectivity: Impacts of L2 age of acquisition and social diversity of language use on

control networks. Neuropsychologia, 117, 123– 134. https:// doi.org/ 10.1016/ j.neuro psyc holo gia.2018.04.037

Hern�ndez, S.E., Suero, J., Barros, A., Gonz�lez- Mora, J.L., & Rubia, K. (2016). Increased grey matter associated

with long- term sahaja yoga meditation: a voxel- based morphometry study. PloS ONE, 11(3), e0150757.

https:// doi.org/ 10.1371/ jour nal.pone.0150 757

Kandel, E.R., Schwartz, J.H., Jessell, T.M., Siegelbaum, S., Hudspeth, A.J., & Mack, S. (Eds.). (2000). Principles

of neural science (vol. 4, pp. 1227– 1246). New York: McGraw- hill.

Klein, D., Mok, K., Chen, J.K., & Watkins, K.E. (2014). Age of language learning shapes brain structure: A cor-

tical thickness study of bilingual and monolingual individuals. Brain and Language, 131, 20– 24. https:// doi.

org/ 10.1016/ j.bandl.2013.05.014

K�hncke, Y., D�zel, S., Sander, M.C., Lindenberger, U., K�hn, S., & Brandmaier, A.M. (2021). Hippocampal

and parahippocampal gray matter structural integrity assessed by multimodal imaging is associated with epi-

sodic memory in old age. Cerebral Cortex, 31(3), 1464– 1477. https:// doi.org/ 10.1093/ cer cor/ bhaa 287

Kousaie, S., & Klein, D. (this volume). Using functional neuroimaging to investigate second language organiza-

tion. In K. Morgan- Short & J.G. van Hell (Eds.), The Routledge handbook of second language acquisition

and neurolinguistics. Routledge.

Korenar, M., & Pliatsikas, C. (this volume). Second language acquisition and neuroplasticity: Insights from

the Dynamic Restructuring Model. In K. Morgan- Short & J.G. van Hell (Eds.), The Routledge handbook of

second language acquisition and neurolinguistics. Routledge.

Kroll, J.F., & Bialystok, E. (2013). Understanding the consequences of bilingualism for language processing and

cognition. Journal of Cognitive Psychology, 25(5), 497– 514. https:// doi.org/ 10.1080/ 20445 911.2013.799 170

Page 12

Eleonora Rossi, Toms Voits, and Vincent DeLuca

Kuhl, P.K., Stevenson, J., Corrigan, N.M., van den Bosch, J.J.F., Can, D.D., & Richards, T.L. (2016).

Neuroimaging of the bilingual brain: Structural brain correlates of listening and speaking in a second lan-

guage. Brain and Language, 162, 1– 9. https:// doi.org/ 10.1016/ j.bandl.2016.07.004

Le Bihan, D., Mangin, J.F., Poupon, C., Clark, C.A., Pappata, S., Molko, N., & Chabriat, H. (2001). Diffusion

tensor imaging: Concepts and applications. Journal of Magnetic Resonance Imaging: An Official Journal

of the International Society for Magnetic Resonance in Medicine, 13(4), 534– 546. https:// doi.org/ 10.1002/

jmri.1076

Leivada, E., Westergaard, M., Du�abeitia, J.A., & Rothman, J. (2021). On the phantom- like appearance of bilin-

gualism effects on neurocognition: (How) should we proceed? Bilingualism: Language and Cognition, 24(1),

197– 210. https:// doi.org/ 10.1017/ S13667 2892 0000 358

Levelt, W. (1989). Speaking: From intention to articulation. MIT Press.

Levelt, W.J. (1993). Speaking: From intention to articulation. MIT press. https:// doi.org/ 10.7551/ mitpr ess/

6393.001.0001

Li, L., Abutalebi, J., Zou, L., Yan, X., Liu, L., Feng, X., … & Ding, G. (2015). Bilingualism alters brain func-

tional connectivity between “control” regions and “language” regions: Evidence from bimodal bilinguals.

Neuropsychologia, 71, 236– 247. https:// doi.org/ 10.1016/ j.neuro psyc holo gia.2015.04.007

Li, P., Legault, J., & Litcofsky, K.A. (2014). Neuroplasticity as a function of second language learning: Anatomical

changes in the human brain. Cortex, 58, 301– 324. https:// doi.org/ 10.1016/ j.cor tex.2014.05.001

Liu, C., Jiao, L., Timmer, K., & Wang, R. (2021). Structural brain changes with second language learning: A

longitudinal voxel- based morphometry study. Brain and Language, 222, Article 105015. https:// doi.org/

10.1016/ j.bandl.2021.105 015

Luk, G., Pliatsikas, C., & Rossi, E. (2020). Brain changes associated with language development and

learning: A primer on methodology and applications. System, 89, Article 102209. https:// doi.org/ 10.1016/

j.sys tem.2020.102 209

Luzzatti, C., & Whitaker, H. (1996). Johannes Schenck and Johannes Jakob Wepfer: Clinical and anatomical

observations in the prehistory of neurolinguistics and neuropsychology. Journal of Neurolinguistics, 9(3),

157– 164. https:// doi.org/ 10.1016/ 0911- 6044(96)00009- 7

Maguire, E.A., Gadian, D.G., Johnsrude, I.S., Good, C.D., Ashburner, J., Frackowiak, R.S., & Frith, C.D. (2000).

Navigation- related structural change in the hippocampi of taxi drivers. Proceedings of the National Academy

of Sciences, 97(8), 4398– 4403. https:// doi.org/ 10.1073/ pnas.070039 597

M�rtensson, J., Eriksson, J., Bodammer, N.C., Lindgren, M., Johansson, M., Nyberg, L., & L�vd�n, M. (2012).

Growth of language- related brain areas after foreign language learning. NeuroImage, 63(1), 240– 244. https://

doi.org/ 10.1016/ j.neu roim age.2012.06.043

Mosso, A. (1881) Ueber den Kreislauf des Blutes im Menschlichen Gehirn. Verlag von Veit & Company. https://

doi.org/ 10.1515/ 978311 2360 606

Pantazis, D., Joshi, A., Jiang, J., Shattuck, D.W., Bernstein, L.E., Damasio, H., & Leahy, R.M. (2010). Comparison

of landmark- based and automatic methods for cortical surface registration. NeuroImage, 49(3), 2479– 2493.

https:// doi.org/ 10.1016/ j.neu roim age.2009.09.027

Patenaude, B., Smith, S.M., Kennedy, D.N., & Jenkinson, M. (2011). A Bayesian model of shape and appearance

for subcortical brain segmentation. NeuroImage, 56(3), 907– 922. https:// doi.org/ 10.1016/ j.neu roim

age.2011.02.046

Pliatsikas, C. (2019). Multilingualism and brain plasticity. In J.W. Schwieter (Ed.), The handbook of the

neuroscience of multilingualism (pp. 230– 251). Wiley- Blackwell. https:// doi.org/ 10.1002/ 978111 9387

725.ch11

Pliatsikas, C. (2020). Understanding structural plasticity in the bilingual brain: The Dynamic Restructuring

Model. Bilingualism: Language and Cognition, 23(2), 459– 471. https:// doi.org/ 10.1017/ S13667 2891

9000 130

Pliatsikas, C., DeLuca, V., Moschopoulou, E., & Saddy, J.D. (2017). Immersive bilingualism reshapes the core

of the brain. Brain Structure and Function, 222(4), 1785– 1795. https:// doi.org/ 10.1007/ s00 429- 016- 1307- 9

Pliatsikas, C., Soares, S.P., Voits, T., DeLuca, V., & Rothman, J. (2021). Bilingualism is a long- term cognitively

challenging experience that modulates metabolite concentrations in the healthy brain. Scientific reports,

11(1), 1– 12. https:// doi.org/ 10.1038/ s41 598- 021- 86443- 4

Rahmani, F., Sobhani, S., & Aarabi, M.H. (2017) Sequential language learning and language immersion in bilin-

gualism: Diffusion MRI connectometry reveals microstructural evidence. Experimental Brain Research,

235(10), 2935– 2945. https:// doi.org/ 10.1007/ s00 221- 017- 5029- x

Raichle, M.E. (2009). A brief history of human brain mapping. Trends in neurosciences, 32(2), 118– 126. https://

doi.org/ 10.1016/ j.tins.2008.11.001

Page 13

Using Structural Neuroimaging to Investigate Second Language

Rosenzweig, M.R., Krech, D., Bennett, E.L., Diamond, M.C. (1962). Effects of environmental complexity

and training on brain chemistry and anatomy: A replication and extension. Journal of Comparative and

Physiological Psychology, 55(4), 429– 437. https:// doi.org/ 10.1037/ h0041 137

Rossi, E., Cheng, H., Kroll, J.F., Diaz, M.T., & Newman, S.D. (2017). Changes in white- matter connectivity in

late second language learners: Evidence from diffusion tensor imaging. Frontiers in Psychology, 8, 2040.

https:// doi.org/ 10.3389/ fpsyg.2017.02040

Rothman, J., Alonso, J.G., & Puig- Mayenco, E. (2019). Third language acquisition and linguistic transfer

(Cambridge Studies in Linguistics). Cambridge University Press. https:// doi.org/ 10.1017/ 978131 6014 660

Roy, C.S., & Sherrington, C.S. (1890) On the regulation of the blood supply of the brain. Journal of Physiology,

11, 85– 108. https:// doi.org/ 10.1113/ jphys iol.1890.sp000 321

Scimeca, M., Carpenter, E., & Kiran, S. (this volume). Aphasia, rehabilitation and second language neurocognition.

In K. Morgan- Short & J.G. van Hell (Eds.), The Routledge handbook of second language acquisition and

neurolinguistics. Routledge.

Stein, M., Winkler, C., Kaiser, A.C., & Dierks, T. (2014). Structural brain changes related to bilingualism: Does

immersion make a difference? Frontiers in Psychology, 5, 1– 7. https:// doi.org/ 10.3389/ fpsyg.2014.01116

Tu, L., Niu, M., Pan, X., Hanakawa, T., Liu, X., Lu, Z., … & Huang, R. (2021). Age of acquisition of Mandarin

modulates cortical thickness in high- proficient Cantonese– Mandarin bidialectals. Journal of Psycholinguistic

Research, 50, 723– 736. https:// doi.org/ 10.1007/ s10 936- 020- 09716- 5

Vaughn, K.A., Nguyen, M.V., Ronderos, J., & Hernandez, A.E. (2021). Cortical thickness in bilingual and mono-

lingual children: Relationships to language use and language skill. NeuroImage, 243, Article 118560. https://

doi.org/ 10.1016/ j.neu roim age.2021.118 560

Weekes, B.S., Abutalebi, J., Mak, H.K.F., Borsa, V., Soares, S.M.P., Chiu, P.W., & Zhang, L. (2018). Effect of

monolingualism and bilingualism in the anterior cingulate cortex: a proton magnetic resonance spectroscopy

study in two centers. Letras de Hoje, 53, 5– 12. https:// doi.org/ 10.15448/ 1984- 7726.2018.1.30954