ArticlePDF Available

A Systematic Review and Meta-Analysis of the Cognitive Correlates of Bilingualism

June 2010
Review of Educational Research 80(2):207-245

June 2010
80(2):207-245

DOI:10.3102/0034654310368803

Authors:

Olusola O. Adesope

Washington State University

Tracy Lavin

Charles Ungerleider

University of British Columbia

A number of studies have documented the cognitive outcomes associated with bilingualism. To gain a clear understanding of the extent and diversity of these cognitive outcomes, the authors conducted a meta-analysis of studies that examined the cognitive correlates of bilingualism. Data from 63 studies (involving 6,022 participants) were extracted and analyzed following established protocols and procedures for conducting systematic reviews and guidelines for meta-analysis. Results indicate that bilingualism is reliably associated with several cognitive outcomes, including increased attentional control, working memory, metalinguistic awareness, and abstract and symbolic representation skills. Overall mean effect sizes varied from small to large, depending on the cognitive outcomes measured, and were moderated by methodological features of the studies.

Summary of coded studies and associated effect sizes

…

Overall weighted mean effect size

…

continued)

…

Figures - uploaded by Charles Ungerleider

Content may be subject to copyright.

Content uploaded by Charles Ungerleider

Content may be subject to copyright.

Content uploaded by Olusola O. Adesope

Content may be subject to copyright.

http://rer.aera.net

Review of Educational Research

DOI: 10.3102/0034654310368803

2010; 80; 207 REVIEW OF EDUCATIONAL RESEARCH

Olusola O. Adesope, Tracy Lavin, Terri Thompson and Charles Ungerleider

Bilingualism

A Systematic Review and Meta-Analysis of the Cognitive Correlates of

http://rer.sagepub.com/cgi/content/abstract/80/2/207

The online version of this article can be found at:

Published on behalf of

http://www.aera.net

http://www.sagepublications.com

can be found at:Review of Educational Research Additional services and information for

http://rer.aera.net/cgi/alerts Email Alerts:

http://rer.aera.net/subscriptions Subscriptions:

http://www.aera.net/reprintsReprints:

http://www.aera.net/permissionsPermissions:

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

207

Review of Educational Research

June 2010, Vol. 80, No. 2, pp. 207–245

DOI: 10.3102/0034654310368803

A Systematic Review and Meta-Analysis of the

Cognitive Correlates of Bilingualism

Olusola O. Adesope

Washington State University

Tracy Lavin and Terri Thompson

Directions Evidence and Policy Research Group

Charles Ungerleider

University of British Columbia

A number of studies have documented the cognitive outcomes associated with

bilingualism. To gain a clear understanding of the extent and diversity of

these cognitive outcomes, the authors conducted a meta-analysis of studies

that examined the cognitive correlates of bilingualism. Data from 63 studies

(involving 6,022 participants) were extracted and analyzed following estab-

lished protocols and procedures for conducting systematic reviews and

guidelines for meta-analysis. Results indicate that bilingualism is reliably

associated with several cognitive outcomes, including increased attentional

control, working memory, metalinguistic awareness, and abstract and sym-

bolic representation skills. Overall mean effect sizes varied from small to

large, depending on the cognitive outcomes measured, and were moderated

by methodological features of the studies.

Keywords: bilingual, cognitive correlates, biliteracy, meta-analysis, systematic

review.

Early research on bilingualism warned that bilingualism could be deleterious to

learning. These early studies concluded that monolingual students outperformed

bilingual students on a range of cognitive tasks (for reviews, see Bhatia & Ritchie,

2006; Hakuta, 1986; Macnamara, 1966).

In a seminal article, Peal and Lambert (1962) introduced the concept of “bal-

anced bilinguals” and demonstrated the methodological weaknesses of previous

bilingual studies, providing a new approach for research on bilingualism. Peal and

Lambert noted that early studies on the effects of bilingualism did not properly

match bilingual and monolingual participants along several dimensions, including

socioeconomic status (SES), second language proficiency (pseudobilingualism),

language of assessment, gender, age, and urban–rural contexts. They noted that

these and other factors may have confounded earlier results showing bilingual

disadvantages on cognitive measures. Controlling for these extraneous factors,

Peal and Lambert found that bilingual participants significantly outperformed

monolinguals on several measures of verbal and nonverbal intelligence.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

208208

Since Peal and Lambert’s (1962) original studies, a considerable body of evi-

dence has accumulated suggesting that bilingualism confers a number of cognitive

benefits. For example, researchers have observed that bilinguals may have greater

metalinguistic awareness (Bialystok, 1987, 1988, 2001b; Diaz, 1985; Diaz &

Klinger, 1991; Ferdman & Hakuta, 1985; Goetz, 2000; Hakuta, 1990; Huber &

Lasagabaster, 2000; Ricciardelli, 1993; Titone, 1997) and enhanced metacognitive

skills (Duncan, 2005). Bilinguals may have stronger symbolic representation and

abstract reasoning skills (Bamford & Mizokawa, 1990, 1992; Berguno & Bowler,

2004; Chan, 2005; Diaz, 1985; Goncz, 1988; Johnson, 1991; McLeay, 2003), as

well as better learning strategies (Bochner, 1996; Ponomarev, 1992). Bilinguals

may also have enhanced problem-solving skills because of their ability to selec-

tively attend to relevant information and disregard misleading information

(Bamford & Mizokawa, 1991; Bialystok, 1999, 2001a, 2005; Bialystok &

Majumber, 1998; Duncan, 2005; Stephens, 1997) and may be able to use this

selectivity to succeed at theory-of-mind tasks, which require the ability to attribute

the behavior of others to their own distinct beliefs, desires, and intentions (Chan,

2005; Goetz, 2000). Bilinguals may have enhanced creative and divergent thinking

skills (Braccini & Cianchi, 1993; Ho, 1987; Konaka, 1997; Ricciardelli, 1993;

Srivastava, 1991) and greater cognitive flexibility (Hakuta, 1990; Iannaccone,

Fraternali, & Vaccia, 1992; Kovacs & Teglas, 2002; Kozulin, 1999).

Although many studies have documented advantages for bilinguals on cogni-

tive tasks, other studies have reported negative, null, or mixed effects of bilingual-

ism (Macnamara, 1966; Rosenblum & Pinker, 1983). To make sense of these

conflicting findings, the current work synthesizes the available research on the

cognitive correlates of bilingualism. This review does not investigate the effective-

ness of bilingual education programs because previous reviews, meta-analyses,

and best-evidence syntheses have addressed the question of program effectiveness,

albeit with inconclusive results (Baker & de Kanter, 1981; Rossell & Baker, 1996;

Slavin & Cheung, 2005; Willig, 1985, 1987). Specifically, the current review

focuses on examining the cognitive correlates of bilingualism and the associated

effect sizes. The following section discusses potential relationships between bilin-

gualism and various cognitive skills.

Attentional Control

There is considerable evidence that bilingual speakers are more readily able

to control their attention while engaged in linguistic and nonverbal tasks com-

pared to monolingual learners (Bialystok, 2001a; Bialystok, Craik, Klein, &

Viswanathan, 2004; Bialystok, Craik, & Ryan, 2006; Emmorey, Luk, Pyers, &

Bialystok, 2008). Several explanations have been advanced for this cognitive

advantage. A dominant perspective suggests that the regular use of two lan-

guages requires that bilinguals control their attention and select the target lan-

guage. Some researchers have claimed that the ability to selectively attend to

different representations may be responsible for the greater attentional control

exhibited by bilingual participants in many studies (Bialystok, 2001a; Bialystok,

Martin, & Viswanathan, 2005; Yoshida, 2008). In other words, these researchers

speculate that bilingual learners’ ability to concurrently hold two languages in

the mind, resisting intrusions of words and grammar from one language into the

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

209209

other, might explain the greater control reflected by improved performance on

tasks with conflicting or distracting information.

More recently, researchers have also shown that the cognitive control of attention

found in studies with bilingual children appears to be sustained into adulthood. For

example, Bialystok et al. (2004) found that adults who have been bilingual since

childhood are more capable than comparable monolingual adults of managing

their attention when presented with tasks requiring cognitive control.

In addition, there is a growing body of evidence that bilingualism may help

offset some age-related cognitive declines by building cognitive reserves that slow

the aging process for adults (Bialystok, Craik, & Freedman, 2007; Bialystok et al.,

2004). In a recent study on the effect of lifelong bilingualism on age-related cogni-

tive decline, Bialystok et al. (2007) found that bilingual adults showed symptoms

of dementia 4 years later than comparable monolinguals, even when other factors

remained constant. The preliminary findings in the literature suggest that “the

lifelong experience of managing two languages attenuates the age-related decline

in the efficiency of inhibitory processing” (Bialystok et al., 2004, p. 301).

Working Memory

There are at least two contrasting hypotheses about the relationship between

bilingualism and working memory. First, the need to manage two languages con-

currently could place greater demands on working memory. This hypothesis sug-

gests that bilingualism may impede efficient processing of information in working

memory because of the cognitive load imposed on working memory (Lee, Plass,

& Homer, 2006; Sweller & Chandler, 1994; van Merrienboer & Sweller, 2005).

Conversely, bilinguals’ well-developed ability to inhibit one language while using

the other may increase the efficiency of their working memory capacity because

working memory resources are properly managed through such inhibitory process-

ing (Bialystok et al., 2004; Bialystok, Craik, & Luk 2008; Fernandes, Craik,

Bialystok, & Kreuger, 2007; Just & Carpenter, 1992; Michael & Gollan, 2005;

Rosen & Engle, 1997).

Research on these competing hypotheses has yielded inconclusive findings

with results depending on the nature of the task (Bialystok et al., 2008). On tasks

that require greater attentional control, bilinguals appear to have greater working

memory capacity than monolinguals (Engle, 2002; Kane, Bleckley, Conway, &

Engle, 2001). In attention-aided tasks, however, the bilingual advantage disappears

(Yang, Yang, Ceci, & Wang, 2005).

Metalinguistic Awareness

Metalinguistic awareness is the ability to think about language. It is the explicit

awareness of linguistic forms and structures and an understanding of how these

relate to and produce meaning (Cazden, 1974). It is hypothesized that the experi-

ence of acquiring and maintaining two different languages—with different forms

and structures—allows bilingual speakers to develop an explicit and articulated

understanding of how language works. For example, bilingual speakers have two

different words for most concepts. Reflecting on this can point to the insight that

words are only arbitrarily and symbolically related to their underlying concepts

(e.g., knowing that dog and chien are concepts for dog makes it obvious that the

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

210

word dog is only an arbitrary symbol). Similarly, when syntactic rules differ across

languages, bilingual speakers of those languages may notice the differences and

become explicitly aware of the syntactic rules—which most monolingual speakers

will know only implicitly. For example, English–Japanese bilinguals may note that

objects follow verbs in English sentences (e.g., “I like chocolate”) but objects

precede verbs in Japanese sentences (e.g., “I chocolate like”). Noting this distinc-

tion provides insight into the specific grammatical rules in each language as well

as into the universal properties of human language.

Over the past decades, researchers have investigated the effects of bilingualism

on children’s metalinguistic development. The majority of studies have found that

bilingual speakers, particularly those highly proficient in both languages, demon-

strate greater metalinguistic awareness than their monolingual counterparts

(Bialystok, Majumder & Martin, 2003; Campbell & Sais, 1995; Galambos &

Hakuta, 1988).

Metacognitive Awareness

Metacognitive awareness refers to knowledge about one’s own cognitive pro-

cesses. It is an awareness of one’s own learning strategies and the mental activities

required to self-regulate the learning process (Flavell, 1978). The process of learn-

ing the vocabulary, syntax, phonology, and morphology of more than one lan-

guage, as well as learning how to use this body of knowledge in contextually

appropriate fashion, may provide bilingual speakers special insight into their own

cognitive processes and learning strategies (Kemp, 2007). Research comparing the

metacognitive awareness of bilinguals and monolinguals is scant but has generally

found that bilinguals show greater metacognitive awareness than monolinguals

(Ransdell, Barbier, & Niit, 2006; Vorstman, De Swart, Ceginskas, & Van Den

Bergh, 2009).

Abstract or Symbolic Reasoning and Creative and Divergent Thinking

Across a number of studies, bilinguals have shown enhanced skills with respect

to creative and divergent thinking and to abstract and symbolic reasoning. In an

investigation on creativity and bilingualism, Ricciardelli (1992) found that bilin-

guals outperformed monolinguals in 20 of the 24 studies reviewed, showing a clear

positive relationship between bilingualism and creativity or divergent thinking.

Peal and Lambert (1962) suggested that bilingual children develop greater cogni-

tive flexibility and creativity as a result of switching between two languages and

two different perspectives. As well, Cummins (1976) has proposed that bilingual-

ism spurs the development of abstract and symbolic reasoning through the experi-

ence of having two different words for most concepts. This helps bilingual children

understand that the relationship between words and their referents is entirely arbi-

trary and represents an abstract symbolic relationship.

Problem Solving

Bilinguals also show evidence of enhanced problem-solving skills, particu-

larly on tasks requiring executive control (i.e., planning, cognitive flexibility,

abstract thinking, rule acquisition, initiating appropriate actions and inhibiting

inappropriate actions, and selecting relevant sensory information; Baddeley,

1996). A bilingual advantage has been demonstrated using the Simon task, the

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

211

dimensional change card sort task, and other similar tasks used to assess exec-

utive control for problem-solving tasks (Bialystok, 1999, 2006). Simon tasks

refer to a family of tasks typically used to investigate interference effects.

In the Simon task, stimuli are presented with different target features and in

different positions. For example, participants may be asked to indicate the

color of either a red or a green square presented on one side of the screen by

pressing a left or a right key. The general finding in the Simon task is that reac-

tion times are slowed when the spatial location of the target and its response

coding do not correspond (incongruent condition) versus when spatial location

and response coding correspond (congruent condition). An incongruent trial

occurs when a signal is presented to the right but its color requires a left-hand

button press. Conversely, signals that require a left-hand response and are also

presented on the left side are referred to as congruent trials. Typically, reaction

times are slower to incongruent compared to congruent trials, a finding referred

to as the congruency effect or interference effect.

The enhanced problem-solving ability may be because of the cognitive flexibil-

ity associated with bilingualism. Because bilinguals have the capacity to choose

between two languages, they may develop more flexibility with respect to thinking

that can be applied to solve problems.

Purpose of the Study

Since Peal and Lambert’s (1962) seminal work, a number of studies have docu-

mented the positive cognitive correlates of bilingualism; however, the magnitude

of these effects remains unclear. The majority of recent studies have shown posi-

tive effects of bilingualism, but some studies have shown that bilingual perfor-

mance is relatively impaired on some cognitive tasks. A few studies have also

demonstrated mixed effects of bilingualism on performance on cognitive tasks.

The current analysis is an attempt to synthesize these results. The goal of the pres-

ent study is to meta-analyze research on the cognitive outcomes of bilingualism by

estimating the effects of bilingualism on specific cognitive measures. Specifically,

the meta-analysis addresses the following research questions:

1. What are the cognitive correlates associated with bilingualism?

2. How do the effects of bilingualism vary when cognitive outcomes are mea-

sured in different geographical locations, in different settings, and at differ-

ent educational levels?

3. How are effect sizes influenced by different combinations of languages spo-

ken by bilinguals?

4. Are the effect sizes influenced by methodological features of the research?

Method

Selection Criteria

To capture all relevant studies on the cognitive benefits of bilingualism, specific

criteria for inclusion were developed. Studies were deemed eligible if:

a. Bilingual participants were reported to be equally (or almost equally) profi-

cient in two languages. Thus, participants who were learning second

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

212

languages were not regarded as bilinguals but rather as second language

learners. Studies with such second language learners were excluded from

this meta-analysis. Bilinguals with learning disabilities or other cognitive

disabilities were excluded.

b. They had an experimental group of bilingual participants and a control group

of monolingual participants.

c. Measured outcomes (cognitive benefits) were clearly reported. These

include attentional control, problem-solving skills, creative and divergent

thinking, cognitive flexibility, learning strategies, symbolic representation

and abstract reasoning skills, metalinguistic awareness, metacognitive skills,

and working memory. We excluded studies that measured only psychosocial

outcomes such as employability or social problem solving and other activi-

ties such as code switching, cross-language priming, and social identity.

d. They reported sufficient data to allow for effect size calculations. When

basic descriptive statistics were not included in a study, other statistics were

used (e.g., t and F statistics), but we coded for reviewer’s confidence in

effect size derivation. Studies with insufficient data for effect size calcula-

tions were excluded.

e. They were publicly available, either online or in library archives.

For multiple reports of the same study (e.g., dissertation, conference proceed-

ings, and journal article), the version published as a journal article was coded, but

in some cases other versions of the published article (e.g., conference proceedings)

were used to make the coding features more extensive and accurate.

Location and Selection of Studies

A comprehensive and systematic search was conducted in the following elec-

tronic databases to locate appropriate studies: Academic Search Premier, Education

Full Text, ERIC (including British and Australian ERIC), Linguistic and Language

Behavior Abstracts, PsycINFO, and Web of Science. A primary search was con-

ducted utilizing Boolean combinations of the controlled vocabulary within each

database for the terms immigrant students, bilingualism, and cognition. A manual

search of the reference lists of earlier reviews of the literature on bilingualism (e.g.,

Bialystok, 2002; Cenoz, 2003) was subsequently conducted.

A total of 5,185 articles were obtained from the search procedure. Two researchers

reviewed the titles, abstracts, and keywords of these articles for possible inclusion by

applying the selection criteria stated above. When abstracts did not contain sufficient

information to determine inclusion or exclusion, the full text of the article was obtained

and read. Duplicate studies were removed, and articles that did not meet the selection

criteria were excluded. Interrater agreement was computed to determine the reliability

of including or excluding articles based on reading only the abstracts. This yielded a

Cohen’s kappa of .88. Researchers discussed all disagreements until they were fully

resolved. A total of 157 articles were retained for secondary screening.

Two researchers independently read the full texts of each of the 157 articles

retained after first inclusion to further determine their suitability based on the

specified criteria for inclusion. Only 39 articles met the second inclusion criterion,

and data from these articles were extracted using EPPI-Reviewer, an online appli-

cation for managing and conducting systematic reviews (Thomas & Brunton,

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

213

2006). Coded variables were organized into 11 major categories in the database. These

include (a) study identification, (b) study characteristics and measured outcomes,

sampling strategy, (g) characteristics of samples in the study, (h) recruitment and

consent, (i) data collection, (j) data analysis, and (k) results and conclusion. In

cases where some variables were not explicitly stated in the study, reviewers made

reasonable inferences and noted the absence of explicit information. The appendix

shows the coding book containing a description of all variables coded under each

category.

Although there are many variants of bilingualism (early bilinguals, late bilin-

guals, balanced bilinguals, etc.), sufficient information was not provided in many

of the studies to code this variable. Nevertheless, as highlighted earlier in the arti-

cle, all of the studies that met our inclusion criteria had participants who were

proficient in two languages before the start of each study.

Some articles reported multiple studies. Hence, 63 studies with an overall sam-

ple size of 6,022 participants were reported in all 39 articles and were included for

meta-analysis. Another interrater reliability analysis was conducted to determine

agreement among researchers on inclusion or exclusion judgments based on full-

text review of all 63 studies, yielding a Cohen’s kappa of .92. Again, researchers

discussed all disagreements until they were fully resolved.

Throughout the design and implementation of this review, guidelines for meta-

analysis provided by Lipsey and Wilson (2001) were followed. The coding scheme

prevented inappropriately combining statistically dependent comparisons in cal-

culating mean effect sizes. To generate a single distribution of effect sizes, a mean

effect size was obtained for each set of statistically dependent effect sizes by aver-

aging over different cognitive outcomes and study characteristics.

Extraction and Calculation of Effect Sizes

Effect size is a standardized metric obtained by calculating the difference

between the means of the experimental (bilingual) and control (monolingual)

groups divided by the pooled standard deviation of the two groups. Hedges (1981)

observed that estimates may yield inflated effect sizes when samples are small. To

correct for such bias in effect size estimation, especially with small sample sizes

(Lipsey & Wilson, 2001, p. 48), the obtained Cohen’s d values were converted to

Hedges’s g, an unbiased estimate (Hedges & Olkin, 1985, p. 81) of the standard-

ized mean difference effect size. When other statistics such as F or t were provided,

these were also used to derive effect sizes or, in some cases, to verify the obtained

d (Cooper & Hedges, 1994).

Data Analysis

Standard meta-analytic guidelines and equations were followed in all data analyses

(Cooper & Hedges, 1994; Hedges & Olkin, 1985; Lipsey & Wilson, 2001). All data

analyses were conducted using Comprehensive Meta-Analysis Version 2.2.048

(Borenstein, Hedges, Higgins, & Rothstein, 2008) and SPSS Version 16.0 for Windows.

Aggregating Effect Sizes

To aggregate effect sizes, the inverse variance weight was computed for each

finding. An aggregate effect size was then obtained from the weighted effect sizes

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

214

to derive an overall weighted mean estimate of the effect of the treatment.

This allowed more weight to be assigned to studies with larger sample sizes. The

standard error of Hedges’s unbiased estimate of the mean effect size was then

computed.

A 95% confidence interval was computed around each weighted mean effect

size to determine statistical significance. Confidence intervals spanning a range

above zero were interpreted as indicating a statistically detectable result favoring

bilinguals with respect to the associated cognitive outcome.

An important aspect of meta-analysis involves the determination of whether the

various effect sizes that are averaged into a mean value all estimate the same pop-

ulation effect size. This assumption of homogeneity of variance was tested by the

Q statistic. When all findings share the same population value, Q has an approxi-

mate chi-square distribution with k – 1 degrees of freedom, where k is the number

of effect sizes or studies for a particular subset. When Q exceeded the critical value

of the chi-square distribution (i.e., p < .05), the mean effect size was reported to be

significantly heterogeneous, meaning that there was more variability in the effect

sizes than would be expected from sampling error and suggesting that each effect

size did not estimate a common population mean (Lipsey & Wilson, 2001). The I2

statistic is reported as a complement to interpret the result of the homogeneity test

(Higgins & Thompson, 2002; Huedo-Medina, Sánchez-Meca, Marín-Martínez, &

Botella, 2006).1

Results

After resolving statistical independence, 63 studies (from 39 articles) were

analyzed.

Table 1 shows a summary of the variables coded for each of the 63 studies. This

includes the study, grade level of participants, total number of participants involved

in each study, languages spoken by the bilingual participants, cognitive benefits

measured, and unbiased effect size, Hedges’s g. In this and subsequent tables,

positive effect sizes show bilingual advantages whereas negative effect sizes show

an advantage for monolinguals over bilinguals on cognitive measures.

For outlier analysis, we used the Comprehensive Meta-Analysis software to

determine the effect of removing a number of effect sizes from the distribution of

effect. The forest plot of the 63 standardized mean difference effect sizes for the

cognitive benefits of bilingualism was examined, and 3 potential outlying studies

were removed. The recalculated results did not increase the fit of the remaining

effect sizes to a simple model of homogeneity (g = 0.39; Qtotal(59) = 287.61, p < .001;

I2 = 79.50%). Because the removal of potential outliers did not produce a homoge-

neous model, a decision was made not to remove any effect sizes from the original

distribution.

Overall Relationship Between Cognitive Outcomes and Bilingualism

Table 2 shows the overall weighted mean and homogeneity statistics of all sta-

tistically independent effect sizes. Table 2 and subsequent tables include the num-

ber of participants (N) in each category, the number of findings (k), the weighted

mean effect size (g) and its standard error (SE), the 95% confidence interval around

the mean, a test of the null hypothesis (z), a test of heterogeneity (Q) with its

215

TABLE 1

Summary of coded studies and associated effect sizes

Study

Grade

range N

Languages

(bilingual)aDominant cognitive benefits measured

Effect

size (g)

Abu-Rabia and Siegel (2002) 4–7 63 Arabic–English Working memory for normally developed 0.20

Abu-Rabia and Siegel (2002) 4–7 31 Arabic–English Working memory for reading disabled –0.23

Bialystok (1997) PreK 81 French–English ASR: Symbolic representation of printb1.04*

Bialystok (1997) PreK 87 Chinese–English ASR: Symbolic representation of print 1.32*

Bialystok (1997) PreK 81 French–English ASR: Symbolic representationc0.07

Bialystok (1997) PreK 87 Chinese–English ASR: Symbolic representation –0.26

Bialystok (2006) PSec 97 Mixed–English Working memory using Simon tasks 0.51*

Bialystok (1999) PreK 30 Chinese–English Attentional Controld0.82*

Bialystok (1999) PreK 30 Chinese–English Attentional Controle0.69

Bialystok, Craik, et al. (2005) PSec 29 French–English Attentional Control 0.63

Bialystok, Craik, and Ruocco (2006)f, Exp. 1 PSec 48 Mixed–English Working memoryg0.48

Bialystok, Craik, and Ruocco (2006), Exp. 2 PSec 48 Mixed–English Working memoryh–0.30

Bialystok et al. (2004), Exp. 1 PSec 20 Tamil–English Attentional control: MRT for younger adultsi2.25*

Bialystok et al. (2004), Exp. 1 PSec 20 Tamil–English Attentional control: MRT for older adults 1.03*

Bialystok et al. (2004), Exp. 2 PSec 64 Tamil–English Attentional control: MRT for younger 2.63*

Bialystok et al. (2004), Exp. 2 PSec 30 Mixed–English Attentional control: MRT for older adults 1.35*

Bialystok et al. (2004), Exp. 2 PSec 20 French–English Attentional control: MRT for older adults 2.45*

Bialystok, Luk, et al. (2005b) 1 132 Mixed–English Metalinguistic awarenessj0.95*

Bialystok and Majumder (1998) K–3 71 Mixed–English Problem solvingk0.35

Bialystok et al. (2003), Exp. 1 K–3 72 French–English Metalinguistic awareness –0.02

Bialystok et al. (2003), Exp. 2 K–3 75 French–English Metalinguistic awareness –0.45

Bialystok et al. (2003), Exp. 3 K–3 64 Chinese–English Metalinguistic awarenessl–0.92*

Bialystok et al. (2003), Exp. 3 K–3 58 Spanish–English Metalinguistic awareness 0.51

Bialystok and Martin (2004), Exp. 1 PreK 67 Cantonese–English Attentional control using the DCCS taskm0.35

Bialystok and Martin (2004), Exp. 2 PreK 30 French–English Attentional control using manual DCCS task 0.83*

Bialystok and Martin (2004), Exp. 3 PreK 53 Chinese–English Attentional control: color shape and color object 0.63*

Bialystok and Martin (2004), Exp. 3 PreK 53 Chinese–English Attentional control: functional location and place 0.42

Bialystok, McBride-Chang, et al. (2005), Exp. 1 K–1 67 Chinese–English Metalinguistic awareness –0.04

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

216

Study

Grade

range N

Languages

(bilingual)aDominant cognitive benefits measured

Effect

size (g)

Bialystok, McBride-Chang, et al. (2005), Exp. 2 K–1 67 Chinese–English Metalinguistic awareness –0.30

Bialystok and Senman (2004), Exp. 2 PreK 95 Mixed–English Attentional controln1.59*

Bialystok and Shapero (2005), Exp. 1 K–1 48 Mixed–English ASR: Reversibility of ambiguous figureso1.14*

Bialystok and Shapero (2005), Exp. 2 K 53 Mixed–English Attentional control using different tasksp0.18

Bruck and Genesee (1995) K 137 French–English Metalinguistic awareness –0.44*

Campbell and Sais (1995) PreK 30 Italian–English Metalinguistic awarenessq0.64

Chen et al. (2004), Exp. 1 2 125 Canton-Mandarin Metalinguistic awareness: tone, onset and rime tasks 0.20

Chen et al. (2004), Exp. 1 4 150 Canton-Mandarin Metalinguistic awareness: tone, onset and rime tasks –0.30

Chen et al. (2004), Exp. 2 1 94 Canton-Mandarin Metalinguistic awareness tasksr0.36

Chen et al. (2004), Exp. 2 2 89 Canton-Mandarin Metalinguistic awareness: tone, onset and rime tasks 0.49*

Chen et al. (2004), Exp. 2 4 106 Canton-Mandarin Metalinguistic awareness: tone, onset and rime tasks 0.06

Clarkson and Galbraith (1992) 6 96 Pidgin–English Problem solving with mathematical tests 0.40

Cromdal (1999) K 54 Swedish–English Metalinguistic awareness taskss0.71*

Davidson and Tell (2005), Exp. 1 PreK 40 Urdu–English ASR: Mutual exclusivity in naming whole objectst0.18

Davidson and Tell (2005), Exp. 1 K–1 40 Urdu–English Mutual exclusivity in naming whole objects 1.22*

Davidson and Tell (2005), Exp. 2 K–1 20 Urdu–English Mutual exclusivity in naming whole objects 0.34

Demie and Strand (2006) 10 1026 Mixed–English Problem solving with a standardized test 0.23*

Demont (2001) 1 43 German-French Metalinguistic awareness: several tasksu0.56

Eviatar and Ibrahim (2000) K–1 76 Russian-Hebrew Metalinguistic awareness using several tasksv0.96*

Galambos and Goldin-Meadow (1990) PreK–1 64 Spanish–English Metalinguistic awareness—correcting errors 0.39

Galambos and Goldin-Meadow (1990) PreK–1 64 Spanish–English Metalinguistic awareness—explaining errors 0.09

Gutierrez-Clellen et al. (2004) 3 44 Spanish–English Working memory using two tasksw–0.09

Humphreys and Mumtaz (2001) 2–3 120 Urdu–English Working memory 1.52*

Jordà-Maria (2003) PSec 160 Catalan-Castilian Metalinguistic awareness: Metapragmatic 0.66*

Keshavarz and Astaneh (2004) PSec 60 Mixed–PersianxMetalinguistic awareness 0.56*

Loizou and Stuart (2003) 1 32 Greek–English Metalinguistic awareness using several tasksy0.96*

Loizou and Stuart (2003) 1 36 English-Greek Metalinguistic awareness using several tasks –0.34

Love et al. (2003) PSec 71 Mixed–English Metalinguistic awareness: syntactic processing 0.20

Merriman and Kutlesic (1993) K–2 78 Serbian–English Metalinguistic awareness 0.48*

Oller et al. (2007) 2 288 Spanish–English Metalinguistic awarenessz1.07*

TABLE 1 (continued)

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

217

Study

Grade

range N

Languages

(bilingual)aDominant cognitive benefits measured

Effect

size (g)

Oller et al. (2007) 4 296 Spanish–English Metalinguistic awareness with several measures 0.49*

Ransdell et al. (2006) PSec 137 Mixed–English Metacognitive awarenessaa 0.16

Reichard and Mokhtari (2004) PSec 350 Mixed–English Metacognitive awareness and strategies 0.39*

Sanz (2000) 8–12 201 Catalan-Spanish Metalinguistic awareness with several measures 0.68*

Sheng et al. (2006) K–3 24 Mandarin–English Metalinguistic awareness with several measuresbb 0.22

Note. PreK = prekindergarten; PSec = postsecondary; N = number of participants; ASR = abstract and symbolic representation; MRT = mean reaction time.

a. The language spoken by the monolingual group is presented last in the bilingual coupling (e.g., Arabic–English bilingual means that English was the language of the monolingual group).

b. Moving word problem was used to determine abstract and symbolic representation (ASR) for literacy.

c. Consistent and inconsistent word size problems were used to assess participants’ understanding of the principles of representing spoken language in printed text.

d. Postswitch items in the dimensional change card sort (DCCS) task were used for this sample.

e. Knowledge items in the DCCS task were used for this sample.

f. Results from 24 monolingual and 24 balanced full bilingual participants were used. We discarded results from the 24 unbalanced bilinguals because they reported minimal use of the second language. For example,

“a participant reporting moderate knowledge of Italian but the opportunity to use it only several times per month when visiting relatives was considered an unbalanced bilingual” (Bialystok et al., 2006, p. 1971).

Experiment 1 was conducted with younger participants and Experiment 2 with older participants.

g. A dual-task classification paradigm was used in both experiments.

h. Although the overall result here shows a bilingual disadvantage for adults, the breakdown of outcome measures actually shows bilingual superiority in letters or numbers (LN) conditions, but overall that advantage

was subdued by a much larger bilingual disadvantage in the animals or musical instruments (A-M) task.

i. Mean reaction times (MRT) for both congruent and incongruent trials on Simon task. Note that MRT was reversed to reflect the inverse relationship between RT and benefit. MRT was categorized as a cognitive

benefit closely associated with attentional control.

j. Chinese–English, Hebrew–English, and Spanish–English bilinguals were compared to English monolinguals on phoneme counting measure.

k. French–English and Bengali–English bilinguals were compared to English monolinguals on block design and water-level tasks to examine whether balanced bilinguals carry over their linguistic advantage into

nonlinguistic tasks.

l. This includes the phoneme segmentation task with first and second graders.

m. This was measured by a computerized DCCS task measuring inhibition of attention—categorized as a cognitive benefit closely associated with attentional control.

n. Appearance–reality tasks were used.

o. Outcome measures derived by using children’s embedded figures task and ambiguous figures task, categorized as abstract and symbolic representation.

p. Outcome measures for inhibition of attention include opposite worlds task and computerized DCCS.

q. MLA (metalinguistic awareness) outcomes include letter detection task.

r. Tone, onset, and rime awareness tasks were each assessed with two pholonogical tasks: the oddity task and the same–different task.

s. Outcome measures include symbol substitution, sentence judgment, and correction tasks.

t. Familiar and unfamiliar objects with or without spare parts were used as outcome measures in all experiments. These measures were categorized as abstract and symbolic representation.

u. This article was written in French. Outcomes used to measure linguistic awareness include the grammatical judgment and correction tasks.

v. Outcome measures include the arbitrariness test, initial and final phoneme tests, and phoneme–syllable deletion test.

w. Outcome measures are the competing language processing task and dual processing comprehension task.

x. Turkish–Persian and Armenian–Persian bilinguals were compared to Persian monolinguals using controlled productive ability test at 2,000- and 3,000-word levels.

y. Measures include the rhyme oddity, syllable completion, cluster onset oddity, single phoneme onset oddity, phoneme elision, and initial phoneme identification tasks.

z. Several metalinguistic awareness measures used including passage comprehension, proofreading, verbal analogies, and oral vocabulary.

aa. Metacognitive awareness was measured by self-ratings of reading, writing, speaking, and listening skills.

bb. Paradigmatic and syntagmatic Trials 1, 2, and 3.

*p < .05.

TABLE 1 (continued)

218

associated degrees of freedom (df), and the percentage of variability that is attribut-

able to true heterogeneity, that is, over and above the sampling error (I2).

Table 2 shows that the overall weighted mean effect size is moderate (g = 0.41)

but with substantial variability among studies (Qtotal = 362.62, p < .001).

Heterogeneity among the full set of studies was anticipated as different studies

measured different cognitive outcomes and there was no reason to expect similar

effect sizes for different outcomes. Following up on this heterogeneity, separate

analyses were conducted for each category of cognitive outcomes.

It was observed that 30 of the 63 independent effect sizes were obtained from

studies conducted by Bialystok and colleagues. A sensitivity analysis was per-

formed so as to investigate the potential bias of including about 47% of the entire

studies conducted by a single principal investigator. Table 2 shows statistically

detectable mean effect sizes irrespective of whether studies were authored by

Bialystok or not. There was an overlap in the confidence intervals across the two

categories. Hence, a decision was made to combine the data from both categories

in subsequent analyses.

Cognitive Correlates of Bilingualism

Table 3 shows the weighted mean effect sizes for different cognitive outcomes

associated with bilingualism. All the outcome measures produced statistically detect-

able mean effect sizes in favor of bilingualism. Attentional control produced the largest

effect with a weighted mean effect size of .96 across 14 studies. Although all the cogni-

tive outcomes in Table 3 had statistically detectable mean effect sizes, most of the effect

size distributions were highly heterogeneous, indicating that the variability among

effect sizes was greater than that expected from sampling error.

Moderator analyses were conducted to investigate this heterogeneity, but given

the small number of studies within some of the categories of cognitive outcomes,

a decision was made to collapse across categories with similar outcomes. We col-

lapsed studies that investigated metalinguistic and metacognitive awareness into

one group, and abstract and symbolic representation, attentional control, and prob-

lem solving were collapsed into another group. We did not include working mem-

ory studies with either of the recategorized groups because the dependent variables

used to measure working memory were markedly different from those used in

studies subsumed under the other two groups. Given the small number of studies

TABLE 2

Overall weighted mean effect size

Effect size

95% confidence

interval

Test

of null Test of heterogeneity

N k g SE Lower Upper z Q df p I2 (%)

All 6,022 63 0.41 0.03 0.36 0.46 15.00* 362.62 62 .00 82.90

Not conducted

by Bialystok

4,245 33 0.38 0.03 0.31 0.44 11.54* 141.26 32 .00 77.35

Conducted by

Bialystok

1,777 30 0.48 0.05 0.39 0.58 9.74* 218.17 29 .00 86.71

*p < .05.

219

TABLE 3

Weighted mean effect sizes for bilingual studies by outcome constructs

Effect size

95% confidence

interval

Test

of null Test of heterogeneity

N k g SE Lower Upper z Q df p I 2 (%)

Outcome constructs (cognitive

outcomes)

Metalinguistic awareness 2,813 29 0.33 0.04 0.26 0.41 8.44* 151.56 28 .00 81.53

Metacognitive awareness 487 2 0.32 0.09 0.14 0.50 3.47* 1.26 1 .26 20.80

Working memory 451 7 0.48 0.10 0.29 0.67 4.90* 41.36 6 .00 85.49

Abstract and symbolic representation 484 8 0.57 0.10 0.39 0.76 6.04* 42.10 7 .00 83.37

Attentional control 594 14 0.96 0.09 0.79 1.13 10.86* 73.30 13 .00 82.26

Problem solving 1,193 3 0.26 0.07 0.13 0.38 3.96* 0.65 2 .72 0.00

Collapsed outcome constructs

Metalinguistic and metacognitive

awareness

3,300 31 0.33 0.04 0.26 0.41 9.14* 152.61 30 .00 80.34

Abstract and symbolic representation,

attentional control, and problem

solving

2,271 25 0.52 0.05 0.43 0.61 11.33* 157.69 24 .000 84.78

*p < .05.

Adesope et al.

220

in the working memory category, moderator analyses were not conducted on this

category. However, we note that bilingualism was associated with greater working

memory, resulting in a moderate effect size of .48. Henceforth, the results section

deals with the two new categories (metalinguistic or metacognitive awareness and

attention and representation).

Table 3 shows weighted mean effect sizes for the recategorized outcome con-

structs. The new categories produced statistically detectable mean effect sizes, and

significant heterogeneity was observed in both categories. In subsequent analyses,

we examined the different moderators that may help explain the variability within

each of these two categories.

Moderator Variable Analyses

A mixed-effects model was used for all moderator variable analyses. A mixed-

effects model uses a random-effects model to combine studies within subgroups and a

fixed-effect model to combine studies across subgroups to yield an overall effect

(Borenstein, Hedges, Higgins, & Rothstein, 2009). By using the random-effects model

to combine studies within subgroups in moderator analyses, a mixed-effects model

typically allows for population parameters to vary across studies, reducing the proba-

bility of committing a Type I error, and is usually regarded as a more rigorous meta-

analytical model than a fixed-effects model only (Borenstein et al., 2009; Denson,

2009; Hedges & Vevea, 1998; National Research Council, 1992).

Relationship Between Bilingualism and

Metalinguistic or Metacognitive Awareness Across

Different Locations, Educational Levels, Settings, Language Groups,

SES, and Region

Table 4 shows weighted mean effect sizes for metalinguistic or metacognitive

awareness outcomes under various conditions. When disaggregated by the geographi-

cal location of the research, the effect of bilingualism was statistically detectable across

studies conducted in the United States, Europe, and the Middle East.

Because the total between-levels variance was statistically detectable, QB(5) =

14.47, p = .01, further analyses showed that studies conducted in Europe were sig-

nificantly different from those conducted in other geographical locations. Although

studies conducted in the United States and the Middle East produced a statistically

detectable effect, they were not significantly different from those conducted in China

or Canada and/or those that are part of the various or mixed category.2

In Table 4, a majority of the studies were conducted with primary school stu-

dents in kindergarten through third grade. A mean effect size was statistically

detectable only for these early primary-level students, possibly because of the

small number of studies with students at other levels.

Classroom studies in which learning activities contributed toward performance

assessment in a program produced statistically detectable effect sizes along with

studies that did not specify the setting. The classroom studies were significantly

different from studies conducted in the laboratory. However, the certainty of this

interpretation is limited by the high number of studies that did not specify the set-

ting under which those studies were undertaken.

Bilingualism was reliably associated with greater metalinguistic or metacogni-

tive awareness when bilinguals spoke Spanish and English or a mix of language

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

221

TABLE 4

Weighted mean effect sizes for studies investigating bilinguals’ metalinguistic and metacognitive awareness under various conditions

Effect size

95% confidence

interval

Test of

null

Test of

heterogeneity

N k g SE Lower Upper z QBdf p

Geographical locations 14.47 5 .01

Canada 672 8 –0.08 0.15 –0.37 0.20 –0.57

United States 757 5 0.55 0.18 0.19 0.90 2.99*

Europe 556 7 0.57 0.17 0.24 0.89 3.41*

Middle East 136 2 0.77 0.30 0.18 1.35 2.56*

China 564 5 0.15 0.18 –0.19 0.49 0.87

Mixed 615 4 0.26 0.20 –0.12 0.65 1.34

Educational level 1.99 4 .74

Preschool 30 1 0.64

Primary (K–3) 1,739 21 0.26 0.11 0.04 0.48 2.33*

Intermediate (4–7) 552 3 0.09 0.27 –0.45 0.63 0.33

Secondary (8–12) 261 2 0.63 0.35 –0.05 1.31 1.80

Postsecondary 718 4 0.36 0.24 –0.11 0.83 1.50

Setting 10.54 2 .01

Classroom 1,232 6 0.65 0.16 0.34 0.96 4.11*

Laboratory 1,359 16 0.07 0.10 –0.13 0.27 0.72

Not specified 709 9 0.42 0.14 0.15 0.70 2.99*

Language of the monolingual group 2.99 3 .56

English 2,042 18 0.22 0.11 0.01 0.44 1.97*

Romance (Castilian, French, and Spanish) 468 4 0.58 0.23 0.13 1.04 2.51*

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

222

TABLE 4 (continued)

Effect size

95% confidence

interval

Test of

null

Test of

heterogeneity

N k g SE Lower Upper z QBdf p

Chinese 564 5 0.15 0.20 –0.24 0.55 0.76

Othera226 4 0.50 0.25 0.02 0.99 2.03*

Languages of the bilingual group 16.52 3 .00

French–English 284 3 –0.32 0.22 –0.74 0.11 –1.44

Spanish–English 770 5 0.54 0.17 0.21 0.87 3.24*

Chinese–English 918 10 0.08 0.12 –0.16 0.32 0.69

Mixedb1,328 13 0.51 0.11 0.29 0.72 4.57*

Socioeconomic status 5.99 3 .11

Medium 485 4 –0.04 0.22 –0.47 0.39 –0.19

High 313 4 0.28 0.23 –0.17 0.74 1.23

Mixed 681 4 0.72 0.22 0.28 1.16 3.21*

Not reported 1,821 19 0.27 0.11 0.07 0.48 2.60*

Region 3.05 3 .39

Urban or inner city 552 8 0.28 0.18 –0.07 0.63 1.59

Suburban 225 3 0.00 0.27 –0.54 0.54 0.01

Rural 76 1 0.96

Not reported 2,447 19 0.30 0.11 0.09 0.51 2.83*

Note. QB = QBetween.

a. The “other” subset refers to studies in which participants spoke languages other than English, Spanish, French, and Mandarin or Chinese. Specifically, monolingual control

participants in the “other” subset spoke Swedish (Cromdal, 1999), Hebrew (Eviatar & Ibrahim, 2000), Persian (Keshavarz & Astaneh, 2004), or Greek (Loizou & Stuart, 2003).

b. The “mixed” subset refers to studies in which participants spoke various language couplings apart from French–English, Spanish–English, and Chinese–English.

*p < .05.

Cognitive Benefits of Bilingualism

223

not including French–English or Chinese–English. Similarly, bilinguals showed a

significant metalinguistic or metacognitive advantage over monolinguals for most

monolingual groups except Chinese-speaking monolinguals.

Many of the studies did not report the SES of participants involved in the studies in

Table 4. Statistically detectable mean effect sizes were found among participants

with mixed SES and those studies that did not report the SES of participants.

However, mean effect sizes did not differ statistically among the four SES catego-

ries (i.e., middle, high, mixed, and not reported).

Relationship Between Bilingualism and Attention and

Representation Across Different Locations, Educational Levels,

Settings, Language Groups, SES, and Region

Table 5 shows weighted mean effect sizes for studies investigating bilinguals’

abstract and symbolic representation, attentional control, and problem-solving

outcomes under various conditions. Studies conducted in Canada and various geo-

graphical locations (mixed) produced statistically detectable mean effect sizes.

The mean effect size for studies conducted in different locations was much higher

than those conducted specifically in the United States, Canada, and Europe.

Studies conducted with preschool, primary, and postsecondary students pro-

duced statistically detectable mean effect sizes, with bilingual postsecondary stu-

dents showing the largest mean effect size (g = 1.76). Post hoc comparisons were

conducted because the total between-levels variance for educational level was sig-

nificant, QB(4) = 12.92, p = .01. Results showed that studies conducted with par-

ticipants from preschool up to 12th grade were not significantly different from one

another but that they were all significantly different and produced lower mean

effect sizes than studies conducted with postsecondary students.

There were no classroom studies in which participants engaged in learning

activities that contributed toward performance assessment. Most of the studies did

not report the setting in which they were conducted. Nevertheless, studies con-

ducted in the laboratory and those that did not report the setting both produced

statistically detectable effect sizes.

Except Urdu–English bilinguals (who were included in only three studies), all

categories of language pairs spoken by bilinguals showed statistically detectable

mean effect sizes, indicating that bilingualism is reliably associated with better atten-

tion and representation skills. Because the total between-levels variance was statisti-

cally detectable, QB(4) = 11.95, p = .02, further analyses showed that studies conducted

with bilinguals who spoke Tamil–English were significantly different from those

conducted with Chinese–English bilinguals and those in the “mixed” category.

In Table 5, statistically detectable mean effect sizes were found among partici-

pants with medium SES and those studies that did not report the SES of partici-

pants. Mean effect sizes did not differ statistically among the four SES categories

(i.e., low, middle, mixed, and not reported). These findings indicate that bilingual-

ism contributes to cognitive benefits irrespective of the SES of participants. Across

all regional categories, bilinguals showed more positive and statistically detectable

mean effect sizes than monolinguals.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

224

TABLE 5

Weighted mean effect sizes for studies investigating bilinguals’ abstract and symbolic representation, attentional control, and

problem solving under various conditions

Effect size

95% confidence

interval

Test of

null Test of heterogeneity

N k g SE Lower Upper z QBdf p

Geographical locations 12.06 4 .02

Canada 915 16 0.72 0.16 0.40 1.03 4.52*

United States 100 3 0.58 0.38 –0.17 1.33 1.52

Europe 1,026 1 0.23

Papua New Guinea 96 1 0.40

Mixed 134 4 1.92 0.36 1.23 2.62 5.42*

Educational level 12.92 4 .01

Preschool 734 12 0.63 0.18 0.28 0.99 3.47*

Primary (K–3) 232 5 0.64 0.29 0.07 1.21 2.20*

Intermediate (4–7) 96 1 0.40

Secondary (8–12) 1,026 1 0.23

Postsecondary 183 6 1.76 0.30 1.18 2.34 5.93*

Setting 0.30 1 .59

Laboratory 421 8 0.72 0.24 0.25 1.19 2.98*

Not specified 1,850 17 0.88 0.17 0.55 1.20 5.28*

Language of the monolingual group 0.44 1 .51

English 2,175 24 0.85 0.14 0.57 1.12 6.01*

Pidgin English 96 1 0.40

Languages of the bilingual group 11.95 4 .02

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

225

TABLE 5 (continued)

Effect size

95% confidence

interval

Test of

null Test of heterogeneity

N k g SE Lower Upper z QBdf p

French–English 342 7 0.80 0.24 0.33 1.28 3.30*

Chinese–English 407 7 0.55 0.23 0.09 1.01 2.36*

Tamil–English 134 4 1.92 0.35 1.23 2.62 5.47*

Urdu–English 100 3 0.58 0.38 –0.16 1.33 1.53

Mixeda1,288 4 0.63 0.29 0.05 1.20 2.14*

Socioeconomic status 4.40 3 .22

Low 1,026 1 0.23

Medium 716 13 1.12 0.20 0.73 1.52 5.52*

Mixed 100 3 0.58 0.43 –0.26 1.42 1.36

Not reported 429 8 0.56 0.25 0.07 1.06 2.23*

Region 4.99 2 .08

Urban or inner city 356 8 0.57 0.27 0.05 1.09 2.16*

Suburban 472 7 0.57 0.27 0.04 1.10 2.09*

Not reported 1,443 10 1.25 0.24 0.78 1.73 5.23*

a. The “mixed” subset refers to studies in which bilinguals spoke languages other than French–English, Chinese–English, Tamil–English, and Urdu–English. Specifically, bilin-

gual participants spoke Bengali/French–English (Bialystok & Majumder, 1998); Armenia, Hebrew, Russian, Spanish, and a host of other languages (Bialystok & Senman, 2004,

p. 573); Pidgin–English (Clarkson & Galbraith, 1992); or various languages (Demie & Strand, 2006).

*p < .05.

226

Effects of Different Methodological Features

Tables 6 and 7 show how effect sizes varied with research design and imple-

mentation features. The studies were categorized according to the researchers’

confidence in the calculated effect size (rated as medium or high depending on

whether sufficient data were provided to calculate an effect size), the reliability and

validity of instruments used in the studies, and a measure of the level of control for

bias in the studies. An additional source of variance often reported in meta-analyses is

the research designs of the studies under consideration (Abrami & Bernard, 2006).

However, because none of the studies in the current meta-analysis used random

assignment—almost all the studies included were matched quasi-experimental

designs except two studies that did not clearly report the design—we could not

meta-analyze the variations based on research designs. As well, all the analyzed

studies were published, hence precluding further analyses based on whether stud-

ies were published or not but heightening the potential for publication bias.

Table 6 shows the weighted mean effect sizes for metalinguistic or metacogni-

tive awareness studies by different methodological features. High coder confi-

dence in the effect size calculation was associated with statistically detectable

mean effect size, but medium coder confidence was not. Studies in which the reli-

ability of measures was not reported produced a statistically detectable mean effect

size, but studies in which reliability was reported did not. Studies produced a sta-

tistically detectable mean effect size whether they reported validity measures or

not. Studies in which bias was tightly controlled or not controlled produced a sta-

tistically detectable mean effect size, but studies in which biases were loosely

controlled did not.

Table 7 shows the weighted mean effect sizes for representation and attention

studies by different methodological features. High coder confidence in the effect

size calculation was associated with a statistically detectable mean effect size, but

medium coder confidence was not. Mean effect sizes were statistically detectable

whether reliability of the measures used were reported or not. Similarly, mean

effect sizes were statistically significant whether bias was tightly or loosely con-

trolled.

Publication Bias

Researchers have observed that published studies are a biased sample of studies

in a particular domain because research reports are more likely to be published

when significant results are reported (Orwin, 1983; Rosenthal, 1979). Hence, stud-

ies with nonsignificant findings are often either tucked away in file drawers or

reported in the less accessible gray literature. This problem, referred to as the “file-

drawer problem,” becomes apparent in meta-analyses, where there is a tendency to

exclude unpublished and gray literature studies, thereby potentially skewing meta-

analytical findings toward a positive mean effect size. This poses a threat to the

validity of results obtained from any meta-analyses. Researchers have proposed

different methods to examine the validity of results obtained from meta-analyses.

In the current work, publication bias seems to be a potential threat to the validity

of results obtained because all the studies analyzed were published in peer-

reviewed journals. Two statistical tests were performed using Comprehensive

Meta-Analysis software to assess the potential for publication bias in this meta-

analysis. First, a “classic fail-safe N” test was performed to determine the number

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

227

TABLE 6

Weighted mean effect sizes for studies investigating bilinguals’ metalinguistic and metacognitive awareness by different

methodological features

Effect size

95% confidence

interval

Test of

null

Test of

heterogeneity

N k g SE Lower Upper z QBdf p

Confidence in effect size derivation 0.84 1 .36

Medium 238 2 0.58 0.33 –0.06 1.21 1.77

High 3,062 29 0.27 0.09 0.09 0.44 3.02*

Reliability 0.44 1 .51

Not reported 2,028 19 0.33 0.11 0.12 0.55 3.06*

Reported 1,272 12 0.22 0.14 –0.05 0.49 1.57

Validity 0.54 1 .46

Not reported 2,644 26 0.26 0.09 0.08 0.45 2.76*

Reported 656 5 0.44 0.22 0.01 0.86 2.00*

Control for bias in studies 3.11 2 .21

Tightly controlled 1,556 17 0.34 0.12 0.12 0.57 2.99*

Loosely controlled 1,023 11 0.12 0.14 –0.15 0.39 0.86

Not controlled 721 3 0.58 0.25 0.10 1.06 2.35*

*p < .05.

228

of null effect studies needed to raise the p value associated with the average effect

above an arbitrary alpha level (set at a = .05). This test revealed that 3,791 addi-

tional studies would be required to invalidate the overall effect found in this meta-

analysis.

The second test, Orwin’s fail-safe N, was used to estimate the number of

file-drawer studies with null results required to nullify the effects found in this

meta-analysis. Using a criterion trivial level of .05, the fail-safe N was found to be

453 studies, meaning that 453 missing null studies would be needed to bring the

current mean effect size found in this meta-analysis to .05. Researchers have

claimed that meta-analytical results could be interpreted as valid and thus resistant

to the “file-drawer problem” if the fail-safe N reaches the 5k+10 limit (Carson,

Schriesheim, & Kinicki, 1990; Rosenthal, 1979). The results of the two computed

statistical tests suggest that it is unlikely that publication bias poses a significant

threat to the validity of findings reported in the current work because both fail-safe

N values are larger than the 5k+10 limit.

Discussion

The meta-analysis presented here synthesized data from 63 studies to examine

the cognitive correlates of bilingualism and the magnitude of such benefits. We

found a moderate positive overall effect of bilingualism on different cognitive

measures. Nevertheless, significant variability existed among studies, with some

yielding a positive cognitive effect of bilingualism and others a negative cognitive

effect. To more appropriately explain the variability among findings, we examined

the study features that may account for the variable effects. Specifically, this meta-

analysis provided answers to the following research questions.

TABLE 7

Weighted mean effect sizes for studies investigating bilinguals’ abstract and

symbolic representation, attentional control, and problem solving by different

methodological features

Effect

size

95% confidence

interval

Test of

null

Test of

heterogeneity

N k g SE Lower Upper z QBdf p

Confidence in effect

size derivation

0.65 1 .42

Medium 1,180 4 0.59 0.34 –0.07 1.24 1.75

High 1,091 21 0.88 0.15 0.58 1.18 5.69*

Reliability 1.43 1 .23

Not reported 1,779 18 0.94 0.17 0.61 1.27 5.58*

Reported 492 7 0.57 0.26 0.07 1.08 2.21*

Control for bias in

studies

3.11 2 .21

Tightly controlled 1,862 18 0.89 0.16 0.57 1.21 5.43*

Loosely controlled 409 7 0.67 0.26 0.17 1.17 2.61*

*p < .05.

229

What Cognitive Correlates Is Bilingualism Associated With?

Results of this meta-analysis show that bilingualism is positively associated

with a range of cognitive benefits. Specifically, bilinguals were found to outper-

form monolinguals on the combined measures of metalinguistic and metacognitive

awareness (g = 0.33) and on measures of abstract and symbolic representation,

attentional control, and problem solving (g = 0.52). There was, however, signifi-

cant variability in these effect sizes.

These results indicate that the process of acquiring two languages and of simul-

taneously managing those languages—of inhibiting one so the second can be used

without interference—allows bilinguals to develop skills that extend into other

domains. These skills appear to give bilingual speakers insight into the abstract

features of language and into their own learning processes. They also appear to

give bilingual speakers an enhanced capacity to appropriately control and distrib-

ute their attentional resources, to develop abstract and symbolic representations,

and to solve problems.

How Do the Effects of Bilingualism Vary When

Cognitive Outcomes Are Measured in Different

Geographical Locations and Settings, for Different Language Groups,

at Different SESs, and at Different Educational Levels?

Sociocultural attitudes toward bilingualism and the use of particular languages

vary across different countries and communities (Hamers & Blanc, 2000).

Consistent effects associated with geography were not observed in the current

work. For example, Canada and the United States provide rather different contexts

for bilingualism: Canada maintains two official languages, and a number of poli-

cies supporting bilingualism in both official languages are in place (Government

of Canada, 2009). As well, Canada’s multicultural attitude toward immigration

encourages immigrants to maintain their native language while acquiring at least

one of Canada’s official languages (Esses & Gardner, 1996). In contrast, the United

States is officially unilingual and has adopted a “melting pot” rather than multicul-

tural approach to immigration (Ravitch, 1990). Despite these differences, we

observed no consistent differences regarding cognitive correlates of bilingualism

in Canada and the United States. It may be that limiting our inclusion criteria to

cover only studies in which participants were balanced bilinguals eliminated any

potential differences across bilingual speakers in different countries.

Bilinguals who acquire their second language at an early age often master that

second language more fully than those who acquire their second language later in

life (Johnson & Newport, 1989). The evidence reviewed in the current analysis

suggests that earlier, rather than later, acquisition of a second language is also more

likely to be associated with greater metalinguistic and metacognitive awareness.

Although bilingual speakers of all ages demonstrated significant advantages with

respect to representation and attention, only the youngest bilinguals (who, by def-

inition, must have acquired their second language early in life) showed significant

advantages with respect to metalinguistic or metacognitive awareness.

Studies conducted in the classroom yielded a moderately high effect of bilin-

gualism over monolingualism on measures of metalinguistic and metacognitive

awareness, but this effect was not observed in laboratory studies conducted outside

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

230

of the classroom. Because of the high stakes associated with classroom studies in

which testing contributes to students’ performance assessments, it is possible that

bilinguals in such studies exhibited their cognitive skills to the fullest relative to

bilinguals in the laboratory studies in which outcome measures typically do not

contribute to performance assessments.

How Are Effect Sizes Influenced by Different Combinations

of Languages Spoken by Bilinguals?

For most language pairings—including French–English and Chinese–English

pairings—bilinguals outperformed monolinguals on the set of representation and

attention measures that includes abstract and symbolic representations, attentional

control, and problem solving. However, when measuring metalinguistic or meta-

cognitive outcomes, French–English and Chinese–English pairings did not show

a significant advantage over monolinguals. This may be a result of the specific

manners in which metalinguistic or metacognitive outcomes were measured in the

French and Chinese speaker studies. Alternatively, this pattern of results may indi-

cate that any language pairing can yield general cognitive benefits for bilinguals

but specific pairs of languages are necessary for bilinguals to develop metalinguis-

tic or metacognitive advantages. For example, it may be necessary for both lan-

guages to contain certain features for bilinguals to become metalinguistically

aware of those features. This would be in line with the findings from research on

cross-linguistic transfer of reading skills, where it has been observed that the

amount of cross-linguistic transfer is maximized when both languages have alpha-

betic writing systems. As a result, cross-linguistic transfer occurs easily for

Spanish–English bilinguals but not for Mandarin–English bilinguals (Bialystok,

Luk, & Kwan, 2005; Koda & Zehler, 2008).

Are the Effect Sizes of Cognitive Benefits Influenced by

Methodological Features of the Research?

Across different methodological features, the meta-analysis found cognitive advan-

tages of bilingualism. Specifically, on measures of metalinguistic or metacognitive

awareness, statistically detectable benefits of bilingualism were obtained when

researchers’ confidence in calculating effect sizes was high and when biases in studies

were tightly controlled. However, the large number of studies that did not report any

reliability or validity measures somewhat undermines the methodological quality of

the studies we analyzed. On measures of representation and attention, a similar result

was found with researchers’ confidence in calculating effect sizes. A large, statistically

detectable effect size was obtained showing that a high coder confidence in the effect

size was associated with a high mean effect size. Also, statistically detectable effect

sizes were obtained whether reliability of instruments used was reported or not and

irrespective of how biases in studies were controlled.

Conclusion

Although monolingualism is often depicted as normative, the best available

evidence indicates that, around the world, bilingual and multilingual speakers out-

number monolingual speakers (Tucker, 1998). The current work suggests that

bilingualism (and, presumably, multilingualism) is associated with a number of

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

231

cognitive benefits. These findings point to the need for further work investigating

the utility of these benefits in a variety of contexts. For example, cognitive benefits

documented in the current work may be of use to bilingual speakers in classrooms

where the language of instruction is not their native language. As the pace of

immigration to developed countries increases, the incidence of bilingualism and

multilingualism in these countries will also increase—as will the number of sec-

ond language learners in public school classrooms. Although second language

learners often present challenges within the classroom, the current analysis sug-

gests they may also bring a number of advantages. It remains unclear how, in

practice, second language learners and their instructors may capitalize on these

advantages. Further work investigating the cognitive correlates of bilingualism

within educational contexts is required to clarify this issue.

APPENDIX

Codebook on the cognitive benefits of bilingualism: Coded variables

Section A: Identification of studies/

reviewer

A.1 Name of the reviewer A.1.1 Details

A.2 Date of the review A.2.1 Details

A.3 Please enter the details of the paper A.3.1 Author (last name, first name)

A.3.2 Date of publication

A.3.3 Title of the article

A.3.4 Journal, issue number, pages

A.3.5 Credentials—institutional affiliation of

the authors

Section B: General information

B.1 What is the focus of the study? B.1.1 ESL instructional practices

B.1.2 Cognitive benefits of bilingualism

B.2 In what country was the study

conducted?

B.2.1 USA

B.2.2 Canada

B.2.3 United Kingdom

B.2.4 Australia/New Zealand

B.2.5 Other (please specify)

B.3 What is the language of the

monolingual group?

B.3.1 English

B.3.2 Other (please specify)

B.4 What are the languages of the

bilingual group?

B.4.1 (Please list)

B.4.2 (Please list)

B.5 What is/are the cognitive variable(s)

being measured?

B.5.1 Problem-solving skills

B.5.2 Theory-of-mind tasks

B.5.3 Creative and divergent thinking

B.5.4 Cognitive flexibility

B.5.5 Abstract representation and reasoning

skills

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

232

B.5.6 Learning strategies

B.5.7 Metalinguistic awareness

B.5.8 Metacognitive skills

B.5.9 Working memory

B.5.10 Others (Please provide details)

B.6 What is the age range of the

participants?

B.6.1 Grades K–3 (5–8 years old)

B.6.2 Grades 4–7 (9–12 years old)

B.6.3 Grades 8–12 (13–18 years old)

B.6.4 Longitudinal range (Please provide details)

B.7 What is the location of the study? B.7.1 Classroom

B.7.2 Pull-out room (for instance, a resource

room or another room within the school)

B.7.3 Laboratory

B.7.4 Other (please specify)

Section C: Study research questions

C.1 What is the overarching question the

researcher is trying to address? Please

write in authors’ description if there is

one. Elaborate if necessary, but

indicate which aspects are reviewers’

interpretations. Other, more specific

questions about the research questions

and hypotheses are asked later.

C.1.1 Explicitly stated (please specify)

C.1.2 Implicit (please specify)

C.1.3 Not stated/unclear (please specify)

C.2 What is the author’s specific

research question? Research questions

operationalize the overarching

question. Please write in author’s

description if there is one. Elaborate if

necessary, but indicate which aspects

are reviewer’s interpretations.

C.2.1 Explicitly stated (please specify)

C.2.2 Implicit (please specify)

C.2.3 Not stated/unclear (please specify)

C.3 What is the author’s hypothesis?

Research questions or hypotheses

operationalize the aims of the study.

Please write in authors’ description if

there is one. Elaborate if necessary,

but indicate which aspects are

reviewers’ interpretations.

C.3.1 Explicitly stated (please specify)

C.3.2 Implicit (please specify)

C.3.3 Not stated/unclear (please specify)

C.4 What is the theoretical/empirical basis

for this study? Please write in author’s

description if there is one. Elaborate if

necessary, but indicate which aspects

are reviewers’ interpretations.

C.4.1 Explicitly stated (please specify)

C.4.2 Implicit (please specify)

C.4.3 Not stated/unclear (please specify)

APPENDIX (continued)

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

233

Section D: Methods–designs

D.3 What variables are included? D.3.1 Independent variables

List independent moderator variables

D.3.2 Dependent (outcome) variables

List outcome variables

D.4 What measurement tool(s) is/are

used?

D.4.1 Standardized test

Please provide the name of the test if listed

D.4.2 Classroom or teacher developed test

Please describe and give page number

D.4.3 Observation

Please give a description and the page

number

D.4.4 Other

Please describe and give page number

Section E: Methods–groups

E.1 What is the design of the study E.1.1 Nonrandomized with treatment and

control groups. How were the groups

assigned/created?

E.1.2 Repeated measures design (Where the

same sample of individuals is measured in

all of the conditions)

E.1.3 Other (please specify)

E.2 How do the groups differ? E.2.1 Explicitly stated (please specify)

E.2.2 Implicitly stated (please specify)

E.2.3 Not applicable (not more than one group)

E.2.4 Not stated/unclear (please specify)

E.3 Number of groups. For instance, in

studies in which comparisons are made

between groups, this may be the number

of groups into which the dataset is

divided for analysis (e.g., social class,

or form size).

E.3.1 Not applicable

E.3.2 One

E.3.3 Two

E.3.4 Three

E.3.5 Four or more (please specify)

E.3.6 Other/unclear (please specify)

Section F: Method–sampling strategy

F.1 Are the authors trying to produce

findings that are representative of a

given population? Please write in

authors’ description. If authors do not

specify, please indicate reviewers’

interpretation.

F.1.1 Explicitly stated (please specify)

F.1.2 Implicit (please specify)

F.1.3 Not stated/unclear (please specify)

F.2 What is the sampling frame (if any)

from which the participants are chosen?

e.g., telephone directory, electoral

There may be two stages—e.g., first

sampling schools and then classes or

pupils within them.

F.2.1 Not applicable (please specify)

F.2.2 Explicitly stated (please specify)

F.2.3 Implicit (please specify)

F.2.4 Not stated/unclear (please specify)

APPENDIX (continued)

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

234

F.3 Which method does the study use to

select people, or groups of people (from

the sampling frame)? e.g., selecting

people at random, systematically—

selecting for example every 5th person,

etc.

F.3.1 Not applicable (no sampling frame)

F.3.2 Explicitly stated (please specify)

F.3.3 Implicit (please specify)

F.3.4 Not stated/unclear (please specify)

Section G: Sample information

G.1 What was the total number of

participants in the study (the actual

sample)? If more than one group is

being compared, please give numbers

for each group.

G.1.1 Not applicable (e.g., study of policies,

documents, etc.)

G.1.2 Explicitly stated (please specify)

G.1.3 Implicit (please specify)

G.1.4 Not stated/unclear (please specify)

G.2 What is the sex of the individuals in

the actual sample? Please give the

numbers of the sample that fall within

each of the given categories. If

necessary refer to a page number in the

report (e.g., for a useful table). If more

than one group is being compared,

please describe for each group.

G.2.1 Not applicable (e.g., study of policies,

documents, etc.)

G.2.2 Single sex (please specify)

G.2.3 Mixed sex (please specify)

G.2.4 Not stated/unclear (please specify)

G.2.5 Coding is based on: Authors’

description

G.2.6 Coding is based on: Reviewers’ inference

G.3 What is the socioeconomic status of

the individuals within the actual

sample? If more than one group is

being compared, please describe for

each group.

G.3.1 Not applicable (e.g., study of policies,

documents, etc.)

G.3.2 Explicitly stated (please specify)

G.3.3 Implicit (please specify)

G.3.4 Not stated/unclear (please specify)

G.4 What is the ethnicity of the individuals

within the actual sample? If more than

one group is being compared, please

describe for each group.

G.4.1 Not applicable (e.g., study of policies,

documents, etc.)

G.4.2 Explicitly stated (please specify)

G.4.3 Implicit (please specify)

G.4.4 Not stated/unclear (please specify)

G.5 What is known about the special

educational needs of individuals

within the actual sample? (choose all

that apply) Please note whether it was

explicitly stated or implicit.

G.5.1 Normally developing children

G.5.2 Language impaired

G.5.3 Learning disabled

G.5.4 Reading disabled

G.5.5 Late talkers

G.5.6 Intellectual difficulties

G.5.7 Other (please specify)

G.5.8 Not applicable (e.g., study of policies,

documents, etc.)

G.5.9 Not stated/unclear (please specify)

G.6 What are the regional characteristics

of individuals/groups in sample?

G.6.1 Not applicable (please specify)

G.6.2 Urban/inner city

G.6.3 Suburban

G.6.4 Rural

G.6.5 Not stated/unclear (please specify)

APPENDIX (continued)

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

235

G.6.6 Other

G.7 Do the authors describe strategies used

to control for bias from confounding

variables and groups? Please include

information on: Were the groups similar

at the start of the study?

G.7.1 Age (please specify)

G.7.2 Gender (please specify)

G.7.3 Social class (please specify)

G.7.4 Not stated/Unclear

G.7.5 Other (please specify)

G.8 What are additional sample

information/characteristics if any?

G.8.1 Not applicable

G.8.2 Sample was obtained from another

study (specify study)

G.8.3 Other

G.8.4 Not stated/unclear (please specify)

G.9 How many participants left before the

end of the study? If more than one group,

please give numbers for each group.

G.10 If the study involves following

samples prospectively over time, do

authors provide baseline values of key

variables such as those being used as

outcomes and relevant socio-

demographic variables?

G.10.1 Not applicable (e.g., study of policies,

documents, etc.)

G.10.2 Not applicable (not following samples

prospectively over time)

G.10.3 Yes (please specify)

G.10.4 No

Section H: Recruitment and consent

H.1 Which methods are used to recruit

people into the study? e.g., letters of

invitation, telephone contact, face-to-

face contact.

H.1.1 Letter of invitation

H.1.2 Telephone contact

H.1.3 Other (please specify)

H.1.4 Not applicable (please specify)

H.1.5 Not stated/unclear (please specify)

H.2 Were any incentives provided to

recruit people into the study?

H.2.1 Not applicable (please specify)

H.2.2 Explicitly stated (please specify)

H.2.3 Not stated/unclear (please specify)

H.3 Whose consent was sought? Please

comment on the quality of consent if

relevant.

H.3.1 Participants

H.3.2 Parental consent

H.3.3 Other (please specify)

H.3.4 No consent was sought

H.3.5 Not stated/unclear

Section I: Data collection

I.1 Which methods were used to collect

the data? Please indicate all that

apply and give further detail where

possible.

I.1.1 Experimental

I.1.2 Curriculum-based assessment

I.1.3 Focus group

I.1.4 Group interview

I.1.5 One to one interview (face to face or by

phone)

I.1.6 Observation

I.1.7 Self-completion questionnaire

APPENDIX (continued)

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

236

I.1.8 Self-completion report or diary

I.1.9 Exams

I.1.10 Clinical test

I.1.11 Practical test

I.1.12 Psychological test

I.1.13 Hypothetical scenario including

vignettes

I.1.14 School/college records (e.g.,

attendance records, etc.)

I.1.15 Secondary data such as publicly

available statistics

I.1.16 Other documentation

I.1.17 Not stated/unclear (please specify)

I.1.18 Coding is based on: authors’ description

I.1.19 Coding is based on: reviewers’

interpretation

I.2 Who collected the data? Please

indicate all that apply and give further

detail where possible.

I.2.1 Researcher

I.2.2 Head teacher/senior management

I.2.3 Teaching or other staff

I.2.4 Parents

I.2.5 Pupils/students

I.2.6 Governors

I.2.7 LEA/government officials

I.2.8 Other educational practitioner

I.2.9 Other (please specify)

I.2.10 Not stated/unclear

I.2.11 Coding is based on: Authors’ description

I.2.12 Coding is based on: Reviewers’ inference

I.3 Do the authors describe any ways

they addressed the reliability of their

data collection tools/methods? e.g.,

test–retest methods (Where more than

one tool was employed, please provide

details for each.)

I.3.1 Details

I.4 Do the authors describe any ways

they have addressed the validity of

their data collection tools/methods?

e.g., mention previous validation of

tools, published version of tools,

involvement of target population in

development of tools. (Where more

than one tool was employed, please

provide details for each.)

I.4.1 Details

APPENDIX (continued)

(continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

237

Section J: Data analysis

J.1 Which statistical methods, if any,

were used in the analysis? (check all

that apply)

J.1.1 Descriptive

J.1.2 Correlation

J.1.3 Group differences (e.g., t test, ANOVA)

(please specify)

J.1.4 Growth Curve analysis/multilevel

modeling (HLM)

J.1.5 Structural equation modeling (SEM)

J.1.6 Path analysis

J.1.7 Regression

J.1.8 Latent growth curve

J.1.9 Other (please specify)

Section K: Results and conclusion

K.1 What are the results of the study as

reported by authors? Please give

details and refer to page numbers in

the report(s) of the study, where

necessary (e.g., for key tables).

Group means:

SD:

Estimated effect size:

Appropriate SD:

F, t statistic:

Significance:

Inverse variance weight:

K.2 Are there any shortcomings in the

reporting of the data? Please list all

implicit and explicit shortcomings of

the study.

K.2.1 Yes (please specify)

K.2.2 No

K.3 Do the authors report on all

variables they aimed to study as

specified in their aims/research

questions? This excludes variables

just used to describe the sample.

K.3.1 Yes

K.3.2 No (please specify)

K.4 What do the author(s) conclude

about the findings of the study?

Please give details and refer to page

numbers in the report of the study,

where necessary.

K.4.1 Details

Note. ESL = English as a second language; LEA = local educational agencies; HLM = hierarchical linear

modeling. Thanks to the EPPI-Reviewer database system team (Thomas & Brunton, 2006).

APPENDIX (continued)

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

238

Notes

This research was supported by a grant from the Canadian Language & Literacy

Research Network (CLLRNet) and funds from the Canadian Council on Learning

(CCL).

1When the value of Q is less than or equal to the degree of freedom associated with

a subset, I2 is assigned a value of zero. Similarly, negative values of I2 are assigned a

value of zero so that I2 lies between 0% and 100%. A value of 0% indicates no observed

heterogeneity, and larger values show increasing heterogeneity. Higgins and Thompson

(2002) recommend that percentages of around 25% (I2 = .25), 50% (I2 = .50), and 75%

(I2 = .75) should be interpreted to mean low, medium, and high heterogeneity, respec-

tively.

2The “mixed” category consists of studies conducted in multiple locations (e.g.,

Ransdell, Barbier, & Niit, 2006, reported a study conducted in three disparate geo-

graphical locations—Estonia, France, and the United States).

References

References marked with an asterisk indicate articles included in the meta-analysis.

Abrami, P. C., & Bernard, R. M. (2006). Research on distance education: In defense of

field experiments. Distance Education, 27, 5–26.

*Abu-Rabia, S., & Siegel, L. S. (2002). Reading, syntactic, orthographic, and working

memory skills of bilingual Arabic-English speaking Canadian children. Journal of

Psycholinguistic Research, 31, 661–678.

Baddeley, A. D. (1996). Exploring the central executive. Quarterly Journal of

Experimental Psychology, 49A, 5–28.

Baker, K. A., & de Kanter, A. A. (1981). Effectiveness of bilingual education: A review

of the literature. Washington, DC: U.S. Department of Education, Office of Planning,

Budget and Evaluation.

Bamford, K. W., & Mizokawa, D. T. (1990, April). Cognitive development of children

in an additive-bilingual program: The third report. Paper presented at the meeting

of the American Educational Research Association, Boston, MA.

Bamford, K. W., & Mizokawa, D. T. (1991, April). Cognitive and language development in

an additive-bilingual program: Report after four observations. Paper presented at the

meeting of the American Educational Research Association, Chicago, IL.

Bamford, K. W., & Mizokawa, D. T. (1992, April). Spanish-immersion children in

Washington State: Fourth year of a longitudinal study. Paper presented at the meet-

ing of the American Educational Research Association, Seattle, WA.

Berguno, G., & Bowler, D. M. (2004). Communicative interactions, knowledge of a

second language, and theory of mind in young children. Journal of Genetic

Psychology, 165, 293–309.

Bhatia, T., & Ritchie, W. (Eds.). (2006). Handbook of bilingualism. Cambridge, MA:

Blackwell.

Bialystok, E. (1987). Words as things: Development of word concept by bilingual

children. Studies in Second Language Acquisition, 9, 133–140.

Bialystok, E. (1988). Levels of bilingualism and levels of linguistic awareness.

Developmental Psychology, 24, 560–567.

*Bialystok, E. (1997). Effects of bilingualism and biliteracy on children’s emerging

concepts of print. Developmental Psychology, 33, 429–440.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

239

*Bialystok, E. (1999). Cognitive complexity and attentional control in the bilingual

mind. Child Development, 70, 636–644.

Bialystok, E. (2001a). Against isolationism: Cognitive perspectives on second language

research. Selected Proceedings of the Second Language Research Forum, 97–103.

Bialystok, E. (2001b). Bilingualism in development: Language, literacy, and cognition.

Cambridge, UK: Cambridge University Press.

Bialystok, E. (2002). Acquisition of literacy in bilingual children: A framework for

research. Language Learning, 52, 159–199.

Bialystok, E. (2005). Consequences of bilingualism for cognitive development. In J. F.

Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic

approaches (pp. 417–432). New York, NY: Oxford University Press.

*Bialystok, E. (2006). Effect of bilingualism and computer video game experience on

the Simon task. Canadian Journal of Experimental Psychology, 60, 68–79.

Bialystok, E., Craik, F. I. M., & Freedman, M. (2007). Bilingualism as a protection

against the onset of symptoms of dementia. Neuropsychologia, 45, 459–464.

*Bialystok, E., Craik, F. I. M., Grady, C., Chau, W., Ishii, R., Gunji, A., & Pantev, C.

(2005). Effect of bilingualism on cognitive control in the Simon task: Evidence from

MEG. NeuroImage, 24, 40–49.

*Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging,

and cognitive control: Evidence from the Simon task. Psychology & Aging, 19, 290–303.

Bialystok, E., Craik, F. I. M., & Luk, G. (2008). Cognitive control and lexical access

in younger and older bilinguals. Journal of Experimental Psychology: Learning,

Memory and Cognition, 34, 859–873.

*Bialystok, E., Craik, F. I. M., & Ruocco, A. C. (2006). Dual-modality monitoring in

a classification task: The effects of bilingualism and aging. Quarterly Journal of

Experimental Psychology, 59, 1968–1983.

Bialystok, E., Craik, F. I. M., & Ryan, J. (2006). Executive control in a modified anti-

saccade task: Effects of aging and bilingualism. Journal of Experimental Psychology:

Learning, Memory, and Cognition, 32, 1341–1354.

*Bialystok, E., Luk, G., & Kwan, E. (2005). Bilingualism, biliteracy, and learning to

read: Interactions among languages and writing systems. Scientific Studies of

Reading, 9, 43–61.

*Bialystok, E., & Majumder, S. (1998). The relationship between bilingualism and the

development of cognitive processes in problem solving. Applied Psycholinguistics,

19, 69–85.

*Bialystok, E., Majumder, S., & Martin, M. M. (2003). Developing phonological

awareness: Is there a bilingual advantage? Applied Psycholinguistics, 24, 27–44.

*Bialystok, E., & Martin, M. M. (2004). Attention and inhibition in bilingual children:

Evidence from the dimensional change card sort task. Developmental Science, 7,

325–339.

Bialystok, E., Martin, M. M., & Viswanathan, M. (2005). Bilingualism across the lifespan:

The rise and fall of inhibitory control. International Journal of Bilingualism, 9, 103–119.

*Bialystok, E., McBride-Chang, C., & Luk, G. (2005). Bilingualism, language profi-

ciency, and learning to read in two writing systems. Journal of Educational

Psychology, 97, 580–590.

*Bialystok, E., & Senman, L. (2004). Executive processes in Appearance-reality tasks:

The role of inhibition of attention and symbolic representation. Child Development,

75, 562–579.

*Bialystok, E., & Shapero, D. (2005). Ambiguous benefits: The effect of bilingualism

on reversing ambiguous figures. Developmental Science, 8, 595–604.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

240

Bochner, S. (1996). The learning strategies of bilingual versus monolingual students.

British Journal of Educational Psychology, 66, 83–93.

Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2008).

Comprehensive meta-analysis (Version 2.2.048). Englewood, NJ: Biostat.

Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction

to meta analysis. Chichester, UK: Wiley.

Braccini, F., & Cianchi, R. (1993). The influence on some linguistic and cognitive

skills of the early learning of a foreign language. Rassegna Italiana Di Linguistica

Applicata, 25, 53–66.

*Bruck, M., & Genesee, F. (1995). Phonological awareness in young second language

learners. Journal of Child Language, 22, 307–324.

*Campbell, R., & Sais, E. (1995). Accelerated metalinguistic (phonological) aware-

ness in bilingual children. British Journal of Developmental Psychology, 13, 61–68.

Carson, K. P., Schriesheim, C. A., & Kinicki, A. J. (1990). The usefulness of the “fail-

safe” statistic (N) in meta-analysis. Educational and Psychological Measurement,

50, 233–243.

Cazden, C. R. (1974). Play with language and metalinguistic awareness: One dimen-

sion of language experience. Urban Review, 7, 28–29.

Cenoz, J. 2003. The additive effect of bilingualism on third language acquisition: A

review. International Journal of Bilingualism, 7, 71–87.

Chan, K. T. (2005). Chinese-English bilinguals’ theory-of-mind development. Dissertation

Abstracts International, A: The Humanities and Social Sciences, 65 (10-A).

*Chen, X., Anderson, R. C., Li, W., Hao, M., Wu, X., & Shu, H. (2004). Phonological

awareness of bilingual and monolingual Chinese children. Journal of Educational

Psychology, 96, 142–151.

*Clarkson, P. C., & Galbraith, P. (1992). Bilingualism and mathematics learning—

another perspective. Journal for Research in Mathematics Education, 23, 34–44.

Cooper, H. M., & Hedges, L. V. (1994). The handbook of research synthesis. New York,

NY: Russell Sage.

*Cromdal, J. (1999). Childhood bilingualism and metalinguistic skills: Analysis and

control in young Swedish English bilinguals. Applied Psycholinguistics, 20, 1–20.

Cummins, J. (1976). The influence of bilingualism on cognitive growth: A synthesis of

research findings and explanatory hypotheses. Working Papers on Bilingualism, 9, 1–43.

*Davidson, D., & Tell, D. (2005). Monolingual and bilingual children’s use of mutual

exclusivity in the naming of whole objects. Journal of Experimental Child

Psychology, 92, 25–45.

*Demie, F., & Strand, S. (2006). English language acquisition and educational attain-

ment at the end of secondary school. Educational Studies, 32, 215–231.

*Demont, E. (2001). Contribution of early second language learning in the develop-

ment of language awareness and learning to read. International Journal of

Psychology, 36, 274–285.

Denson, N. (2009). Do curricular and co-curricular diversity activities influence racial

bias? A meta-analysis. Review of Educational Research, 79, 805–838.

Diaz, R. M. (1985). The intellectual power of bilingualism. Quarterly Newsletter of the

Laboratory of Comparative Human Cognition, 7, 16–22.

Diaz, R. M., & Klinger, C. (1991). Towards an explanatory model of the interaction between

bilingualism and cognitive development. In E. Bialystok (Ed.), Language processing in

bilingual children (pp. 167–192). Cambridge, UK: Cambridge University Press.

Duncan, S. E. (2005). Child bilingualism and cognitive functioning: A study of four

Hispanic groups. Dissertation Abstracts International, A: The Humanities and

Social Sciences, 65(8), 2895–A.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

241

Emmorey, K., Luk, G., Pyers, J. E., & Bialystok, E. (2008). The source of enhanced

cognitive control in bilinguals. Psychological Science, 19, 1201–1206.

Engle, R. W. (2002). Working memory capacity as executive attention. Current

Directions in Psychological Science, 11, 19–23.

Esses, V. M., & Gardner, R. C. (1996). Multiculturalism in Canada: Context and cur-

rent status. Canadian Journal of Behavioural Science, 28, 145–152.

*Eviatar, Z., & Ibrahim, R. (2000). Bilingual is as bilingual does: Metalinguistic abili-

ties of Arabic-speaking children. Applied Psycholinguistics, 21, 451–471.

Ferdman, B. M., & Hakuta, K. (1985, April). A population perspective on bilingualism

in Puerto Rican children. Paper presented at the meeting of the Society for Research

in Child Development, Toronto, Canada.

Fernandes, M. A., Craik, F. I. M., Bialystok, E., & Kreuger, S. (2007). Effects of bilin-

gualism, aging, and semantic relatedness on memory under divided attention.

Canadian Journal of Experimental Psychology, 61, 128–141.

Flavell, J. H. (1978). Metacognitive development. In J. M. Scandura & C. J. Brainerd

(Eds.), Structural process theories of complex human behavior (pp. 213–245).

Alphen aan den Rijn, Netherlands: Sijthoff & Noordhoff.

*Galambos, S. J., & Goldin-Meadow, S. (1990). The effects of learning two languages

on levels of metalinguistic awareness. Cognition, 34, 1–56.

Galambos, S., & Hakuta, K. (1988). Subject-specific and task-specific characteristics

of metalinguistic awareness in bilingual children. Applied Psycholinguistics, 9,

141–162.

Goetz, P. J. (2000). A comparison of 3- and 4-year-old English monolinguals, man-

darin monolinguals, and mandarin-English bilinguals on a series of theory of

mind-related tasks. Dissertation Abstracts International, B: Sciences and

Engineering, 61, 1110–B.

Goncz, L. (1988). A research study on the relation between early bilingualism and

cognitive development. Psychologische Beitrage, 30, 75–91.

Government of Canada. (2009). Roadmap for Canada’s linguistic duality, 2008–2013:

Acting for the future. Retrieved from http://www.pch.gc.ca/pgm/slo-ols/strat-eng.

cfm

*Gutierrez-Clellen, V. F., Calderon, J., & Weismer, S. E. (2004). Verbal working mem-

ory in bilingual children. Journal of Speech, Language, and Hearing Research, 47,

863–876.

Hakuta, K. (1986). Mirror of language: The debate on bilingualism. New York, NY:

Basic Books.

Hakuta, K. (1990). Bilingualism and bilingual education: A research perspective

(Occasional Papers in Bilingual Education No. 1). Washington, DC: National

Clearinghouse for Bilingual Education.

Hamers, J. F., & Blanc, M. H. A. (2000). Bilinguality and bilingualism (2nd ed.).

Cambridge, UK: Cambridge University Press.

Hedges, L. V. (1981). Distribution theory for Glass’s estimator of effect size and related

estimators. Journal of Educational Statistics, 6, 107–128.

Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. Orlando, FL:

Academic Press.

Hedges, L. V., & Vevea, J. L. (1998). Fixed and random-effects models in meta-analysis.

Psychological Methods, 3, 486–504.

Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-

analysis. Statistics in Medicine, 21, 1539–1558.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

242

Ho, D. Y. (1987). Bilingual effects on language and cognitive development: With spe-

cial reference to Chinese-English bilinguals. Bulletin of the Hong Kong Psychological

Society, 18, 61–69.

Huber, E., & Lasagabaster, D. (2000). The cognitive effects of bilingualism. ITL,

Review of Applied Linguistics, 129–130, 191–224.

Huedo-Medina, T. B., Sánchez-Meca, F., Marín-Martínez, F., & Botella, J. (2006).

Assessing heterogeneity in meta-analysis: I2 or Q statistic? Psychological Methods,

11, 193–206.

*Humphreys, G. W., & Mumtaz, S. (2001). The effects of bilingualism on learning to

read English: Evidence from the contrast between Urdu-English bilingual and

English monolingual children. Journal of Research in Reading, 24, 113–134.

Iannaccone, A., Fraternali, O. T., & Vaccia, R. (1992). Cognitive flexibility and early

bilingual education. Rassegna Italiana Di Linguistica Applicata, 24, 149–165.

Johnson, J. (1991). Constructive processes in bilingualism and their cognitive growth

effects. In E. Bialystok (Ed.), Language processing in bilingual children (pp. 193–

221). Cambridge, UK: Cambridge University Press.

Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language

learning: The influence of maturational state on the acquisition of English as a sec-

ond language. Cognitive Psychology, 21, 60–99.

*Jordà-Maria, P. S. (2003). Metapragmatic awareness and pragmatic production of

third language learners of English: A focus on request acts realizations. International

Journal of Bilingualism, 7, 43–68.

Just, M. A., & Carpenter, P. A. (1992). A capacity theory of comprehension: Individual

differences in working memory. Psychological Review, 98, 122–149.

Kane, M. J., Bleckley, M. K., Conway, A. R. A., & Engle, R. W. (2001). A controlled-

attention view of working-memory capacity. Journal of Experimental Psychology:

General, 130, 169–183.

Kemp, C. (2007). Strategic processing in grammar learning: Do multilinguals use more

strategies? International Journal of Bilingualism, 4, 241–261.

*Keshavarz, M. H., & Astaneh, H. (2004). The impact of bilinguality on the learning

of English vocabulary as a foreign language (L3). International Journal of Bilingual

Education and Bilingualism, 7, 295–302.

Koda, K., & Zehler, A. (Eds.). (2008). Learning to read across languages: Cross-

linguistic relationships in first- and second-language literacy development. New

York, NY: Routledge.

Konaka, K. (1997). The relationship between degree of bilingualism and gender to

divergent thinking ability among native Japanese-speaking children in the New York

Area. Dissertation Abstracts International, AAT 9718715.

Kovacs, A. M., & Teglas, E. (2002). Integrating two languages, theories of minds, and

executive functions. Odense Working Papers in Language and Communication, 3, 1.

Kozulin, A. (1999). Reality monitoring, psychological tools, and cognitive flexibility

in bilinguals: Theoretical synthesis and pilot experimental investigation. In P. Lloyd

& C. Fernyhough (Eds.), Lev Vygotsky: Critical assessments: Future directions (Vol.

4, pp. 187–198). New York, NY: Routledge.

Lee, H., Plass, J. L., & Homer, B. D. (2006). Optimizing cognitive load for learning

from computer-based science simulations. Journal of Educational Psychology, 98,

902–913.

Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA:

Sage.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

243

*Loizou, M., & Stuart, M. (2003). Phonological awareness in monolingual and bilin-

gual English and Greek five-year-olds. Journal of Research in Reading, 26, 3–18.

*Love, T., Maas, E., & Swinney, D. (2003). The influence of language exposure on

lexical and syntactic language processing. Experimental Psychology, 50, 204–216.

Macnamara, J. (1966). Bilingualism and primary education. Edinburgh, UK: Edinburgh

University Press.

McLeay, H. (2003). The relationship between bilingualism and the performance of

spatial tasks. International Journal of Bilingual Education and Bilingualism, 6,

423–438.

*Merriman, W. E., & Kutlesic, V. (1993). Bilingual and monolingual children’s use of

two lexical acquisition heuristics. Applied Psycholinguistics, 14, 229–249.

Michael, E., & Gollan, T. H. (2005). Being and becoming bilingual: Individual differ-

ences and consequences for language production. In J. F. Kroll & A. M. B. de Groot

(Eds.), The handbook of bilingualism: Psycholinguistic approaches (pp. 389–407).

New York, NY: Oxford University Press.

National Research Council. (1992). Combining Information: Statistical issues and

opportunities for research. Washington, DC: National Academy Press.

*Oller, D. K., Pearson, B. Z., & Cobo-Lewis, A. B. (2007). Profile effects in early

bilingual language and literacy. Applied Psycholinguistics, 28, 191–230.

Orwin, R. (1983). A fail-safe N for effect size in meta-analysis. Journal of Educational

Statistics, 8, 157–159.

Peal, E., & Lambert, W. E. (1962). The relation of bilingualism to intelligence.

Psychological Monographs, 76, 1–23.

Ponomarev, S. Y. (1992). Bilingualism and its influence on personality development.

Vestnik Sankt-Peterburgskogo Universiteta, Seriya 2: Istoriya, Yazykoznanie,

Literaturovedenie, 3, 46–49.

*Ransdell, S., Barbier, M., & Niit, T. (2006). Metacognitions about language skill and

working memory among monolingual and bilingual college students: When does

multilingualism matter? International Journal of Bilingual Education and

Bilingualism, 9, 728–741.

Ravitch, D. (1990). Multiculturalism: E pluribus plures. American Scholar, 59, 339–340.

*Reichard, C., & Mokhtari, K. (2004). Investigating the strategic reading processes of first

and second language readers in two different cultural contexts. System, 32, 379–394.

Ricciardelli, L. A. (1992). Creativity and bilingualism. Journal of Creative Behavior,

26, 242–254.

Ricciardelli, L. A. (1993). An investigation of the cognitive development of Italian-

English bilinguals and Italian monolinguals from Rome. Journal of Multilingual and

Multicultural Development, 14, 345–346.

Rosen, V. M., & Engle, R. W. (1997). The role of working memory capacity in retrieval.

Journal of Experimental Psychology: General, 126, 211–227.

Rosenblum, T., & Pinker, S. A. (1983). Word magic revisited: Monolingual and bilin-

gual children’s understanding of the word-object relationship. Child Development,

54, 773–780.

Rosenthal, R. (1979). The “file drawer problem” and tolerance for null results.

Psychological Bulletin, 86, 638–641.

Rossell, C. H., & Baker, K. (1996). The educational effectiveness of bilingual educa-

tion. Research in the Teaching of English, 30, 7–69.

*Sanz, C. (2000). Bilingual education enhances third language acquisition: Evidence

from Catalonia. Applied Psycholinguistics, 21, 23–44.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Adesope et al.

244

*Sheng, L., McGregor, K. K., & Marian, V. (2006). Lexical-semantic organization in

bilingual children: Evidence from a repeated word association task. Journal of

Speech, Language, and Hearing Research, 49, 572–587.

Slavin, R. E., & Cheung, A. (2005). A synthesis of research on language of reading instruc-

tion for English language learners. Review of Educational Research, 75, 247–284.

Srivastava, B. (1991). Creativity and linguistic proficiency. Psycho-Lingua, 21,

105–109.

Stephens, M. A. (1997). Bilingualism, creativity, and social problem-solving. Dissertation

Abstracts International, AAT 9729615.

Sweller, J., & Chandler, P. (1994). Why some material is difficult to learn. Cognition

& Instruction, 12, 185–233.

Thomas, J., & Brunton, J. (2006). EPPI-Reviewer 3.0: Analysis and management of

data for research synthesis. London, UK: Social Science Research Unit, Institute of

Education.

Titone, R. (1997). Early second-language learning: Bilingualism and metalinguistic

development. Europe Plurilingue, 12–13, 138–149.

Tucker, G. R. (1998). A global perspective on multilingualism and multilingual educa-

tion. In J. Cenoz & F. Genesee (Eds.), Beyond bilingualism: Multilingualism and

multilingual education (pp. 3–15). Clevedon, UK: Multilingual Matters.

van Merrienboer, J., & Sweller, J. (2005). Cognitive load theory and complex learning:

Recent developments and future directions. Educational Psychology Review, 17,

147–177.

Vorstman, E., De Swart, H., Ceginskas, V., & Van Den Bergh, H. (2009). Language

learning experience in school context and metacognitive awareness of multilingual

children. International Journal of Multilingualism, 6, 258–280.

Willig, A. (1985). A meta-analysis of selected studies on the effectiveness of bilingual

education. Review of Educational Research, 55, 269–317.

Willig, A. (1987). Examining bilingual education research through meta-analysis

and narrative review: A response to Baker. Review of Educational Research, 57,

363–376.

Yang, H., Yang, S., Ceci, S., & Wang, Q. (2005). Effects of bilinguals’ controlled-

attention on working memory and recognition. In J. Cohen, K. McAlister, K.

Rolstad, & J. MacSwan (Eds.), Proceedings of the 4th international symposium on

bilingualism. Somerville, MA: Cascadilla Press.

Yoshida, H. (2008). The cognitive consequences of early bilingualism. Journal of Zero

to Three, 29, 26–30.

Authors

OLUSOLA ADESOPE holds a Ph.D. in Educational Psychology from Simon Fraser

University, Canada. He recently accepted a position as Assistant Professor of Educational

Psychology in the Department of Educational Leadership & Counseling Psychology at

Washington State University, Pullman, WA 99164-2136; email: adesope@yahoo.com.

His research interests include the cognitive and psychological processes underlying mul-

timedia learning, teaching and learning with interactive concept maps and diagrams,

language pedagogy, technology-mediated instruction and research synthesis, especially

meta-analysis.

TRACY LAVIN is a Researcher with the Directions Evidence and Policy Research Group,

1055 Dunsmuir–Suite 1254, Four Bentall Centre, Vancouver, BC V7X 1A2; email:

lavin.t@gmail.com. She received a Ph.D. from the University of British Columbia and

pursued postdoctoral studies at Northwestern University. She has studied and written

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

Cognitive Benefits of Bilingualism

245

about a number of learning and educational issues from both academic and public policy

perspectives. Her research interests include studying how children and adults learn lan-

guage and acquire its underpinning cognitive concepts. She has written several research

papers for the National Literacy Strategy initiative and also leads the team of analysts

responsible for developing, analyzing, and reporting on Canadian Council on Learning’s

annual Survey of Canadian Attitudes toward Learning.

TERRI THOMPSON is a Researcher with the Directions Evidence and Policy Research

Group, 1055 Dunsmuir–Suite1254, Four Bentall Centre, Vancouver, BC V7X 1A2;

email: territho@gmail.com. Over the past five years, she has participated in, and man-

aged the completion of, over 60 systematic reviews and question scans devoted to issues

about education and learning. She holds a Master’s degree in Education from the

University of British Columbia where she studied issues of social justice and education

policy.

CHARLES UNGERLEIDER is a professor of the sociology of education at the University

of British Columbia, 2125 Main Mall, Vancouver, BC V6T 1Z4, Canada; e-mail: charles.

ungerleider@ubc.ca. His research interests include educational policy, finance and gov-

ernance, educational change, and the educational trajectories of students in elementary

and secondary schools.

at SIMON FRASER LIBRARY on June 28, 2010 http://rer.aera.netDownloaded from

The Complexity in Bilingual Code-Switching Research: A Systematic Review

Article

Full-text available

Jun 2024

This systematic review explored how researchers operationalized bilingualism when investigating the relationship between bilingual code-switching experience and cognition. Through a PRISMA-guided systematic review of thirty-two studies with original data, published in English, focusing on adult non-clinical samples, with bilingualism as a key variable, we aimed to understand the prevalence of these issues. Criteria for inclusion required an assessment of bilingualism beyond language proficiency or age of acquisition, and consideration of naturalistic code-switching behaviors. We report our results through an analysis of themes that included aspects of language that are considered when measuring bilingualism and code-switching experience. We present our findings and offer insights for future research, advocating for the inclusion of sociocultural factors and more complex analytical modeling in bilingualism research to foster an evolution in the field.

Probing the Role of Multilingualism and Working Memory in Cross-Situational Word Learning

Article

Full-text available

Jun 2024

Cross-situational word learning (CSWL), the ability to resolve word–referent ambiguity across encounters, is a powerful mechanism found in infants, children, and adults. Yet, we know little about what predicts individual differences in CSWL, especially when learning different mapping structures, such as when referents have a single name (1:1 mapping structure) or two names (2:1 mapping structure). Here, we investigated how multilingual experience and working memory skills (visuo-spatial and phonological) contributed to CSWL of 1:1 and 2:1 structures. Monolingual (n = 78) and multilingual (n = 106) adults completed CSWL tasks of 1:1 and 2:1 structures, a symmetry span task, and a listening span task. Results from path models showed that multilingualism predicted visuo-spatial working memory but not CSWL. Additionally, phonological working memory predicted accuracy on CSWL of 1:1 structure, but not 2:1 structure. Findings highlight the importance of considering language experience and cognitive skills together to better understand the factors that promote individual CSWL skills.

Bilingüismo y discapacidad intelectual: habilidades lingüísticas y metalingüísticas, en castellano y en catalán, de hablantes con el síndrome Prader-Willi

Article

Jun 2024

Este estudio investiga las habilidades lingüísticas y metalingüísticas de bilingües de castellano y catalán con discapacidad intelectual ligada a síndromes genéticos. A los participantes se les administraron pruebas estandarizadas y una prueba experimental. Respecto a esta última, adoptando el marco cognitivo metalingüístico de Bialystok y colaboradores, ocho bilingües con síndrome Prader-Willi completaron, tanto en castellano como en catalán, una tarea de juicios de gramaticalidad que manipulaba la gramaticalidad y la plausibilidad semántica de los ítems experimentales. Como resultado, la tarea contenía oraciones que incluían diferentes niveles de conocimiento lingüístico y habilidades de control ejecutivo. Trabajos previos reflejan que los bilingües con desarrollo típico muestran (i) resultados similares en ambas lenguas en aquellas oraciones que requieren una alta exigencia de control ejecutivo pero un bajo conocimiento lingüístico (medida metalingüística) y (ii) mejores resultados en la lengua dominante en aquellas oraciones que requieren un alto conocimiento lingüístico pero una baja exigencia de control ejecutivo (medida lingüística). Los resultados obtenidos reflejan que los bilingües incluidos en este estudio muestran habilidades lingüísticas y metalingüísticas comparables en las dos lenguas.

Plurilingual classes: challenges and effects of practices to foster plurilingualism at school

Chapter

Full-text available

Apr 2024

Cultural and linguistic diversity are nowadays a prominent characteristic of schools in most European countries. However, monolingualism is mostly the norm, and even in multilingual countries schools follow a monolingual policy. Thanks to strong empirical evidence bilingualism should no longer be considered a problem or an obstacle to (majority) language learning and academic success. In addition, teachers who are able to implement intercultural and plurilingual pedagogical approaches, doing so with- out negative bias towards ethnic minority students, offer students the best possibilities of academic success, and social and identity integration. In the present article we review the literature about bilingual develop- ment, the sociocultural aspects of ethnically diverse classrooms and discuss the possibility of implementing plurilingual

Exploring the Correlation between English Proficiency and Reading Comprehension among Elementary Learners in the Philippines

Article

Jun 2024

This descriptive-correlational study aimed to determine the relationship between the English proficiency and reading comprehension of the Grade 6 learners at an elementary school in the province of Davao Occidental. Specifically, it aimed to determine the level of English proficiency of the respondents in terms of grammar and vocabulary and their level of reading comprehension. With the use of complete enumeration, there were 20 Grade 6 learners selected as respondents of the study. A validated questionnaire was used to assess the level of students’ English proficiency and the level of students’ reading. The data were analyzed using mean and Pearson-r correlation analysis. Results revealed that the level of English proficiency of the respondents is proficient. This implies that the learners show great understanding and application of the English language. Also, the level of reading comprehension of the respondents is described as high, indicating that they have strong understanding and reading comprehension abilities. Further, the result revealed that there is no significant relationship between the English proficiency and reading comprehension of the Grade 6 learners. It is recommended that further studies on other factors that affect the reading comprehension of elementary learners should be conducted.

Bilingualism and flexibility in task switching: A close replication study

Article

Full-text available

May 2024

This study aimed to closely replicate Wiseheart et al. ( Bilingualism: Language and Cognition , 19(1), 141–146, 2016) by investigating the transferability of language-switching skills to nonlinguistic task switching. Current evidence is mixed and there is a need to conduct robust replications in this area. Bilingual (n = 31) and monolingual (n = 47) young adults characterized stimuli by either colour or shape based on a given cue. Modifications include online data collection (as opposed to in-person) and adapting the nonverbal intelligence test used. All other aspects of the study mirror those by Wiseheart et al. Results indicate that the bilinguals exhibited better cognitive flexibility in task switching, as evidenced by a reduced global switch cost compared with monolinguals. In contrast, mixed evidence was found for local switch costs. Findings mirror those reported by Wiseheart et al. and suggest that by employing comparable task-switch paradigms and recruiting samples matched on several key variables, including age, gender, variety of languages spoken, and use of English, bilingualism does seem to confer broader executive function advantages. Findings are discussed in relation to theoretical implications to inform future replication studies and advance the bilingual advantage in the switching debate.

Self-regulation and comprehension in shared reading: The moderating effects of verbal interactions and E-book discussion prompts

Article

Full-text available

Jun 2024
CHILD DEV

The study examined how children's self‐regulation skills measured by the strengths and weaknesses of ADHD symptoms and normal behavior rating are associated with story comprehension and how verbal engagement and e‐book discussion prompts moderate this relation. Children aged 3–7 ( N = 111, 50% female, Chinese as first language) read an interactive Chinese–English bilingual story e‐book with or without discussion prompts twice with their parents (2020–2021). Results demonstrated that the lower children's self‐regulation skills, the more they struggled with story comprehension. Critically, our data suggest that embedding e‐book discussion prompts and more verbalization in English can mitigate this negative association for children with inattention/hyperactivity. These findings have critical implications for future e‐book design, interventions, and home reading practice for children with inattention/hyperactivity and those at risk for attention deficit/hyperactivity disorder.

Babbling and words: a dynamic systems perspective on phonological development

Chapter

Nov 2015

The most authoritative resource for students and researchers, The Cambridge Handbook of Child Language has been thoroughly updated and extended. Enhancements include new chapters on the acquisition of words, processing deficits in children with specific language impairments, and language in children with Williams syndrome, new authors for the bilingualism and autism chapters, a refocused discourse chapter on written narratives, and a new section on reading and reading disorders, cementing the handbook's position as the best study of the subject available. In a wide-ranging survey, language development is traced from prelinguistic infancy to adolescence in typical and atypical contexts; the material is intuitively grouped into six thematic sections, enabling readers to easily find specific in-depth information. With topics as varied as statistical learning, bilingualism, and the neurobiology of reading disorders, this multidisciplinary Handbook is an essential reference for students and researchers in linguistics, psychology, cognitive science, speech pathology, education and anthropology.

Exploring the Correlation between English Proficiency and Reading Comprehension among Elementary Learners in the Philippines

Article

Jan 2024

Gender Differences in Language Acquisition Patterns among Adolescents in Japan

Article

Full-text available

Jun 2024

Ayaka Tanaka

Purpose: The aim of the study was to analyze the gender differences in language acquisition patterns among adolescents in Japan Methodology: This study adopted a desk methodology. A desk study research design is commonly known as secondary data collection. This is basically collecting data from existing resources preferably because of its low cost advantage as compared to a field research. Our current study looked into already published studies and reports as the data was easily accessed through online journals and libraries. Findings: This study examines gender differences in language acquisition among Japanese adolescents. Males typically prefer informal methods like digital media for vocabulary, while females favor formal instruction. Females show stronger syntax, grammar, and pragmatic skills, driven by a greater interest in language learning for social reasons. Males' attitudes vary more, influenced by socioeconomic factors. Unique Contribution to Theory, Practice and Policy: Social learning theory, gender schema theory & critical feminist theory may be used to anchor future studies on the gender differences in language acquisition patterns among adolescents. By tailoring language instruction to the specific needs of male and female adolescents, educators can create more inclusive learning environments that support equitable language development. Policymakers should prioritize the integration of gender-sensitive language policies and initiatives within educational systems.

Effects of Bilingualism and Biliteracy on Children’s Emerging Concepts of Print

Article

Full-text available

May 1997

Ellen Bialystok

Three groups of 4- and 5-year-old children were examined for their concepts of how print refers to language. All of the children could identify printed letters and their sounds but not read alone. The groups studied were monolingual speakers of English, bilingual speakers of French and English, and bilingual speakers of Chinese (Mandarin) and English. Bilingual children were equally proficient in both languages and were familiar with print and storybooks in both languages. The tasks assessed children’s understanding of the general correspondence between print and language in which the printed form represents a word and the specific correspondence between a constituent of print and one of language that determines representation in a given writing system. The general correspondence relation applies to all writing systems, but the specific correspondence relation changes for different kinds of writing systems. Bilingual children understood better than monolingual children the general symbolic representation of print. The older Chinese–English bilingual children also showed advanced understanding of the specific correspondence relations in English print.

The Role of Working Memory Capacity in Retrieval

Article

Full-text available

Sep 1997

The verbal fluency task requires generation of category exemplars and appears to be an example of what M. Moscovitch (1995) calls a strategic test of memory retrieval. Four experiments explored the role of individual differences in working memory (WM) capacity on verbal fluency under various secondary load conditions. High WM participants consistently recalled more exemplars. However, load conditions caused a decline in recall only for high WM participants. Low WM participants showed no effect of secondary workload on exemplar generation. WM group differences and load effects were observed even in the 1st min of retrieval, which suggests that differences were not due to differences in knowledge. A model of retrieval is supported that relies on cue-based-automatic activation, monitoring of output for errors, controlled suppression of previously recalled items, and controlled strategic search.

A Capacity Theory of Comprehension: Individual Differences in Working Memory

Article

Full-text available

Jan 1992

A theory of the way working memory capacity constrains comprehension is proposed. The theory proposes that both processing and storage are mediated by activation and that the total amount of activation available in working memory varies among individuals. Individual differences in working memory capacity for language can account for qualitative and quantitative differences among college-age adults in several aspects of language comprehension. One aspect is syntactic modularity: The larger capacity of some individuals permits interaction among syntactic and pragmatic information, so that their syntactic processes are not informationally encapsulated. Another aspect is syntactic ambiguity: The larger capacity of some individuals permits them to maintain multiple interpretations. The theory is instantiated as a production system model in which the amount of activation available to the model affects how it adapts to the transient computational and storage demands that occur in comprehension.

A Controlled-Attention View of Working-Memory Capacity

Article

Full-text available

Jun 2001

In 2 experiments the authors examined whether individual differences in working-memory (WM) capacity are related to attentional control. Experiment 1 tested high- and low-WM-span (high-span and low-span) participants in a prosaccade task, in which a visual cue appeared in the same location as a subsequent to-be-identified target letter, and in an antisaccade task, in which a target appeared opposite the cued location. Span groups identified targets equally well in the prosaccade task, reflecting equivalence in automatic orienting. However, low-span participants were slower and less accurate than high-span participants in the antisaccade task, reflecting differences in attentional control. Experiment 2 measured eye movements across a long antisaccade session. Low-span participants made slower and more erroneous saccades than did high-span participants. In both experiments, low-span participants performed poorly when task switching from antisaccade to prosaccade blocks. The findings support a controlled-attention view of WM capacity.

Consequences of Bilingualism for Cognitive Development

Chapter

Feb 2009

Ellen Bialystok

Until recently, cognitive science virtually ignored the fact that most people of the world are bilingual. During the past ten years this situation has changed markedly. There is now an appreciation that learning and using more than one language is the more natural circumstance of cognition. As a result, there is a wealth of new research on second-language learning and bilingualism that provides not only crucial evidence for the universality of cognitive principles, but also an important tool for revealing constraints within the cognitive architecture. In this volume, Judith Kroll and Annette de Groot have brought together the scientists at the forefront of research on second-language learning and bilingualism to present chapters that, rather than focusing simply on their own research, provide the first comprehensive overviews of this emerging field. Bilingualism provides a lens through which each of the central questions about language and cognition can be viewed. The five sections of this book focus on different facets of those questions: How is language acquired when infants are exposed to multiple-language input from birth, and how is it acquired when adults are required to learn a second language after early childhood? How do adult bilinguals comprehend and produce words and sentences when their two languages are potentially always active and in competition with one another? What are the neural mechanisms that underlie proficient bilingualism? What are the general consequences of bilingualism for cognition and for language and thought? This handbook will be essential reading for cognitive psychologists, linguists, applied linguists, and educators who wish to better understand the cognitive basis of bilingualism and the logic of experimental and formal approaches to language science.

Constructive processes in bilingualism and their cognitive growth effects

Chapter

Apr 1991

Janice Johnson

This book is a collection of papers that explore the ways in which bilingual children cope with two language systems. The papers address issues in linguistics, psychology, and education bearing on the abilities that bilingual children use to understand language, to perform highly specialised operations with language, and to function in school settings. All of the papers provide detailed analysis about how specific problems are solved, how bilingualism influences those solutions, and how the social context affects the process. Finally, the implications of these findings for policy-setting and the development of bilingual education programmes are explored. This will be an important and useful volume at the forefront of topical research in an area which is exciting increasing interest among linguists and cognitive scientists.

Bilinguality and Bilingualism

Book

Feb 2000

This updated and revised edition of Hamers and Blanc's successful textbook presents state-of-the-art knowledge about languages in contact from individual bilingualism (or bilinguality) to societal bilingualism. It is both multi- and interdisciplinary in approach, and analyses bilingualism at individual, interpersonal, and societal levels. Linguistic, cognitive and sociocultural aspects of bilingual development are explored, as are problems such as bilingual memory and polyglot aphasia. Hamers and Blanc analyse the relationship between culture, identity, and language behaviour in multicultural settings, as well as the communication strategies in interpersonal and intergroup relations. They also propose theoretical models of language processing and development, which are then applied to bilingual behaviour. Other topics reviewed include language shift, pidgins and creoles, language planning and bilingual education. This book will be invaluable to students, teachers and scholars interested in languages in contact in a range of disciplines including psycholinguistics, linguistics, the social sciences, education and language planning.

The Handbook of Research Synthesis

Article

May 1995

Introduction to meta-analysis

Book

Jan 2009

Die kognitiven Auswirkungen der Zweisprachigkeit

Article

Jan 2000

The relationship between bilingualism and its cognitive effects has always been very controversial. Nevertheless, there is ample evidence in the literature that allows us to state that what seems irrefutable is that bilingualism should not be automatically associated with holding back the subject's cognitive development. To the contrary, under certain conditions, it has been shown that its effects can be clearly beneficial. This research study was completed in the Basque Country, where both Basque and Spanish are official languages and therefore taught at school from kindergarten. However, the availability of three different linguistic models leads to the existence of different degrees of bilingualism regarding Basque students. As a result of this situation, it is hypothesized that the degree of bilingualism, in accordance with the threshold level hypothesis (CUMMINS, 1976), will influence the development of metalinguistic awareness and also will have a significant effect on the English (L3) scores of our sample. It is finally concluded that the degree of bilingualism has a significant positive cognitive effect with regard to the proposed dependent variables

A Systematic Review and Meta-Analysis of the Cognitive Correlates of Bilingualism

Abstract and Figures

Recommended publications

Is Bilingualism Related to a Cognitive Advantage in Children? A Systematic Review and Meta-Analysis

Spatial negative priming in bilingualism

Bilinguismo e controle inibitório: possíveis interferências das variáveis “profissão” e “escolaridad...

Bialystok, Ellen

Bilingualism and enhanced attention in early adulthood