Content uploaded by Martin Jakobus Koch
Author content
All content in this area was uploaded by Martin Jakobus Koch on Mar 08, 2023
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rmmm20
Journal of Multilingual and Multicultural Development
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rmmm20
Extending the scope of the ‘cognitive advantage’
hypothesis: multilingual individuals show higher
flexibility of goal adjustment
Werner Greve, Martin Koch, Verena Rasche & Kristin Kersten
To cite this article: Werner Greve, Martin Koch, Verena Rasche & Kristin Kersten (2021):
Extending the scope of the ‘cognitive advantage’ hypothesis: multilingual individuals show
higher flexibility of goal adjustment, Journal of Multilingual and Multicultural Development,
DOI: 10.1080/01434632.2021.1922420
To link to this article: https://doi.org/10.1080/01434632.2021.1922420
© 2021 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group
Published online: 30 Apr 2021.
Submit your article to this journal
Article views: 1281
View related articles
View Crossmark data
Extending the scope of the ‘cognitive advantage’hypothesis:
multilingual individuals show higher flexibility of goal
adjustment
Werner Greve, Martin Koch, Verena Rasche and Kristin Kersten
Psychology Department, University of Hildesheim, Hildesheim, Germany
ABSTRACT
The cognitive advantage (CA) hypothesis claims that multilingualism
promotes the development of several basic cognitive capacities. A large
number of empirical findings support this hypothesis, but recently there
have also been numerous contradictory findings and methodological
objections. The present paper extends the investigation of possible
cognitive advantages from basic cognitive (executive) functions to
broader cognitive competencies such as cognitive flexibility. A
promising candidate for this is ‘flexibility of goal adjustment’(FGA), a
capacity of developmental regulation that solves problems through
flexible adaptation processes. In a study with N= 119 monolingual and
multilingual adults we found the predicted positive correlation between
multilingualism and FGA. However, the mediator function of executive
capacities entailed in the CA hypothesis operationalised as Stroop and
flanker tasks could not be demonstrated.
ARTICLE HISTORY
Received 29 June 2020
Accepted 20 April 2021
KEYWORDS
Multilingualism; cognitive
advantage; executive
functions; flexibility of goal
adjustment
Although the old conjecture that multilingualism may not only decrease competence in each indi-
vidual language but also harm cognitive functions in general is obviously inaccurate in this simple
form, the thesis that multilingualism has positive consequences for cognitive capabilities of speakers
ist also controversial; interrelationships of these processes are complex (Antoniou 2019). In the pre-
sent paper, we will argue that the discussion could benefit from extending the perspective on poss-
ible cognitive benefits of multilingualism beyond single concrete skills to complex functional groups
of skills, especially cognitive flexibility.
The ‘narrow’cognitive advantage hypothesis: inconclusive findings
The ‘cognitive advantage’(CA) hypothesis claims that multilingualism,
1
due to constant high cog-
nitive demands (e.g. language switching), promotes and enhances not just those cognitive functions
that are particularly related to language (e.g. metalinguistic awareness, Baker and Wright 2017;
Adesope et al. 2010) but at the same time several more basic underlying multi-purpose cognitive
functions such as working memory or information inhibition (Antoniou 2019; Bialystok 2017; Bia-
lystok and Barac 2013; De Groot 2011, 385ff; Valian 2015; Zhang 2018). A number of findings sup-
port this hypothesis (Bialystok 2015; Bialystok et al. 2009; Bialystok, Craik, and Luk 2012; Van den
Noort et al. 2019; for meta-analyses see Adesope et al. 2010; Grundy and Timmer 2017), especially
for those cognitive abilities required to inhibit cognitive content that is constantly activated but not
© 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://
creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the
original work is properly cited, and is not altered, transformed, or built upon in any way.
CONTACT Werner Greve wgreve@uni-hildesheim.de
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT
https://doi.org/10.1080/01434632.2021.1922420
currently needed (Bialystok and Viswanathan 2009; Kroll et al. 2012; Poarch 2018; Poarch and van
Hell 2012), but also for cognitive resources and abilities required in task switching (Bialystok and
Martin 2004; Wisehart, Viswanathan, and Bialystok 2016) and for resources and regulations of
working memory (Grundy and Timmer 2017).
On the other hand, an increasing number of studies did not find support for the CA hypothesis
(e.g. Paap and Greenberg 2013; for recent examples see Czapka, Klassert, and Festman 2019;Czapka
et al. 2020; for overviews see Festman, Czapka, and Winsler 2021; Hilchey and Klein 2011; Paap, John-
son, and Sawi 2015; Poarch and Krott 2019; Van den Noort et al. 2019; Zhou and Krott 2016). The
comprehensive meta-analysis by Lehtonen et al. (2018) concludes that the available evidence does not
provide systematic support for the CA hypothesis concerning executive functions (EFs; e.g. inhi-
bition). The recent large study by Nichols et al. (2020) including a sample of 11,000 people supports
this position. Although the critical stance towards the CA hypothesis has not remained without
response (Titone et al. 2017;forasummaryseeAntoniou2019), the call for more differentiated meth-
odological designs and measurements and more differentiated theoretical explanations with respect to
possible conditions (i.e. moderators; Antoniou and Wright 2017; Festman, Czapka, and Winsler
2021) is certainly reasonable and appropriate (Laine and Lehtonen 2018).
A conclusion of the controversial debate on how the inconsistent results in the empirical litera-
ture can be explained is made difficult (apart from a possible publication bias; de Bruin, Treccani,
and Della Sala 2015) by the fact that the available studies focus on different basic cognitive functions
(executive functions, capacities of working memory, etc.) and, in particular, pursue different meth-
odological approaches with respect to the assessment of the cognitive functions investigated (with
not always comparable methodological rigour; Poarch and Krott 2019; Van den Noort et al. 2019;
Zhou and Krott 2016). In addition, samples are often small (especially in studies which corroborate
the CA hypothesis: Paap, Johnson, and Sawi 2015; Nichols et al. 2020), and are oftentimes not really
comparable (Van den Noort et al. 2019) with regard to age and social characteristics of the mono-
and multilingual subsamples (Czapka et al. 2020; Nichols et al. 2020; Poarch and Krott 2019)and
the involved languages (Antoniou 2019; Bialystok 2017; De Groot 2011; Titone et al. 2017; Valian
2015). Methodological objections extend to some of the available meta-studies on the topic (Paap
et al. 2020). Even the premise of the CA hypothesis that regular switching between languages is cog-
nitively demanding (which in turn leads to the presumed promotion effect) may need to be eval-
uated in a more differentiated way (Blanco-Elorrieta and Pylkkänen 2018), for instance with
respect to typological differences between the respective languages (Antoniou and Wright 2017).
Mental flexibility beyond basic cognitive mechanics: making the case for an
extension of the CA hypothesis
However, the current debate on the CA hypothesis is mainly focused on cognitive functions in a
narrow (‘mechanical’) sense, including in particular executive and memory functions (Nichols
et al. 2020). In contrast, the perspective that advantages of multilingualism may manifest themselves
in more general mental abilities (Fan et al. 2015; Festman and Kersten 2010; Festman and Schwieter
2019; Titone et al. 2017) has received less attention.
The main argument of the present paper claims that the stimulating and hence promoting effect
of multilingualism goes far beyond its ‘narrow’cognitive demands with respect to executive func-
tions. Rather, multilingualism involves a regular change of perspective and focus not only in lin-
guistic terms (grammar, vocabulary, pronunciation), but also in sociolinguistic, socio-cultural
and pragmatic terms, e.g. the social connotations of concepts, the social etiquette of using certain
words, phrases, and idioms, the implicit rules of behaviour associated with them, and the cultural
concepts and associations underlying and intertwined with a linguistic system (‘cultural frame
switching’, Hong et al. 2000; see also Baker and Wright 2017; Kramsch 2004; Pavlenko 2011,
2014). To be able to use such varying perspectives in a competent way is an important part of
advanced multilingual competence. This requires a much more far-reaching mental flexibility
2W. GREVE ET AL.
than simply inhibiting a lexical item or selecting a correct grammatical form or syntactical structure
from the respective linguistic systems (Fan et al. 2015). Rather, multilingual competence might
improve the general ability to flexibly change perspective and focus. For instance, Kovács (2009)
found that bilingual 3-year-old children were better at ‘theory of mind’-tasks (for similar results
with respect to adults: Rubio-Fernandez and Glucksberg 2012). Similarily, Fan et al. (2015)
found that monolingual children failed to interpret a speaker’s meaning more often than both bilin-
gual children and children who were exposed to a multilingual environment but were not bilingual
themselves. Plausibly, such ‘broader’cognitive or mental flexibility envisioned here entails or sup-
ports divergent, creative, elastic and open-ended thinking (Baker and Wright 2017, see p. 142; Fan
et al. 2015; see also Festman and Kersten 2010). A number of studies indicate indeed that multilin-
guals show improved abilities of divergent and creative thinking, e.g. the ability to find numerous
answers to a given problem (Kharkhurin 2007,2009,2015; Lee and Kim 2011). Accordingly, bilin-
gualism predicts individual flexibility, that is, the ability to learn from new experiences or to adjust
behaviour according to contingency (Dewaele and Botes 2019). Given that such abilities are valu-
able or even necessary prerequisites for solving problems and coping with challenges (Adesope et al.
2010), the CA hypothesis can be expanded to the question whether multilingualism may promote
problem-solving competencies and abilities for which the ability to change perspectives and focus
(in a broader sense) is also characteristic or relevant.
Flexibility of goal adjustment as a possible beneficiary of multilingualism
In the present study, we argue that the individual flexibility of goal adjustment (FGA) as a particular
means of solving problems might be a prototypical sample case of such a general ability of problem
solving through using, and flexibly altering and changing, a variety of perspectives. The individual’s
FGA is conceptualised in the framework of the two-process model of developmental regulation
(Brandtstädter 2006,2007,2015; Brandtstädter and Renner 1990; Brandtstädter and Rothermund
2002). According to this model, people regulate problems in two basic ways. In the ‘assimilative’
mode of tenacious goal pursuit, people strive to alter the perceived state of affairs with the available
resources and skills in such a way that the situation fits better to one’s goals or wishes. Within this
mode, the goals pursued remain unaffected. However, if goals appear permanently blocked and
wishes become unattainable, tenacious persistence will not be sufficient to overcome limitations.
Here, the ‘accommodative’mode reduces or resolves problems by adjusting the goals, desires, nor-
mative orientations and preferences to the (perceived) circumstances by devaluing unattainable
goals or desires, replacing them by attainable goals or strivings, or altering one’s standards and
norms. According to the model, the accommodative mode is not one single process but encom-
passes a broad variety of different processes (Brandtstädter 2006; Brandtstädter and Rothermund
2002; Thomsen 2016). This sometimes involves letting go of the previous goal or reorienting
towards a new (reachable) one (e.g. ‘If I cannot become a medical doctor, I may become a salesman
or an artist’), but often means changing the perspective on the previous goal (e.g. reinterpretation:
‘Maybe I can achieve the goal of wanting to help people in a different way than through my failed
medical studies –perhaps as a nurse?’) or broaden the perspective (e.g. downward comparison:
‘What happened to me is depressing and painful –but many other people have been through
much worse things!’). One important –and oftentimes necessary –element of such accommodative
processes can be seen in the change of perspective on the problem.
A number of studies have shown that individuals vary considerably with respect to their prepared-
ness and ability to accommodate their goals or values (for an overview see for example Brandtstädter
2006,2015; Brandtstädter and Rothermund 2002; Heckhausen, Wrosch, and Schulz 2010,2019;
Haase, Heckhausen, and Wrosch 2013). Accordingly, the individual’sFGA(Brandtstädterand
Renner 1990)is a predictor for the stability of the self and well-being (Brandtstädter and Greve
1994), and a buffer against aversive circumstances and experiences both for adolescents (Thomsen
et al. 2015) and adults (e.g. Greve, Leipold, and Kappes 2017; Rühs, Greve, and Kappes 2017).
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 3
The very fact that individual differences can be found indicates that FGA depends on develop-
mental processes and conditions (Meyer and Greve 2012). In order to explain these developmental
preconditions of individual FGA, earlier studies have investigated several conditions (Greve and
Thomsen 2013; Thomsen and Greve 2013; Meyer and Greve 2012) such as individual capabilities
of emotion regulation (Lessing et al. 2017) and social transfer processes (e.g. social learning; Kappes
and Thomsen 2020; Thomsen et al. 2017). For the present context, however, two presumed devel-
opmental conditions for FGA are of particular importance:
(1) Contextual conditions of individual ontogenesis of FGA. Some findings suggest that hetero-
geneous life experiences and developmental stimulations, especially in the first two decades of
life, seem to be related to the development of FGA. Several studies show that heterogeneous leisure
activities and the experience of free and unrestrained play in childhood are linked to individual
FGA (Greve and Thomsen 2013,2016; Greve, Thomsen, and Dehio 2014; Thomsen and Greve
2013). A constraint of these studies, however, is that retrospectively reported childhood experiences
may be influenced by recollection biases. In particular, individuals who are more strongly disposed
to react to challenges in an accommodative manner (i.e. with higher FGA) may also tend to remem-
ber more (and more accurately) unrestrictive play in their childhood (Greve and Thomsen 2016). A
second caveat concerns the fact that leisure activity choices (even in childhood) may also be con-
founded with the individual’s FGA. Moreover, in these studies the heterogeneity of developmental
conditions is limited to one domain of life (leisure). Thus, it would be stronger evidence for the
hypothesis of flexibility-stimulating effects of heterogeneous experiences if biographical infor-
mation can be collected which is, arguably, independent of the individual’s FGA.
In the present study, we claim that multilingualism is a prototypical example for contextual het-
erogeneity. Multilingualism entails heterogeneous ‘cognitive contexts’as multilinguals have, by
definition, to switch between two or more languages and their cultural contexts in which they
are embedded: Multilingualism is, as a rule, associated with cultural diversity (Bialystok 2017; Gros-
jean 2015) and thus entails ‘cultural frame switching’(Hong et al. 2000). For instance, in the case of
migration-related multilingualism, the cause of multilingualism is oftentimes the experience of
more than one socio-cultural environment (Bialystok et al. 2009). At the same time, the self-assess-
ment of multilingualism is not likely to be confounded with FGA. Hence, if multilingualism is used
as a predictor in empirical studies (even in a cross-sectional ones), it could counteract both objec-
tions with respect to the developmental explanation of FGA.
(2) Cognitive conditions of FGA. The ability to think divergently and change perspectives is a
necessary, possibly a constitutive condition to be able to generate alternative goals and relieving per-
spectives on obstacles and problems (such as cognitive reinterpretations or downward comparisons,
Brandtstädter and Rothermund 2002). For instance, adolescents who succeed in looking at tasks
from different perspectives and solving them with creative or divergent thinking show higher
FGA scores than adolescents who tend to approach such tasks in a more rigid manner (Greve
and Thomsen 2013; Thomsen and Greve 2013). If such competencies of divergent thinking rest
on ‘basic’cognitive abilities as referred to in the CA hypothesis (e.g. task switching), it is plausible
to assume that such basic cognitive functions might be supporting conditions for (the development
of) FGA. As to date, studies that have tested this assumption are rare. With respect to preschool
children, the assumption that executive functions could play a direct supporting role in the devel-
opment of accommodative processes could not be confirmed (Lessing et al. 2019; Piekny et al.
2017). However, the development of capabilities of divergent thinking and (as a possible conse-
quence) of FGA might need some more time and become assessable only at a later age. So far, how-
ever, this assumption has not been tested with adults. Thus, the significance of ‘basic’cognitive
functions (EFs) for the development of FGA as mediated through divergent thinking has still to
be investigated.
4W. GREVE ET AL.
Hypotheses
It is the intention of the present paper to investigate a (possible) advantage of multilingual individ-
uals with respect to accommodative self-regulation (FGA). Moreover, we will test the hypothesis
that this relationship is mediated by basic cognitive functions (EFs).
Hypothesis 1: Multilingualism predicts FGA . If the assumption is correct that heterogeneous life
experiences and conditions promote FGA (Greve and Thomsen 2013,2016; Thomsen and Greve
2013), then the heterogeneity of linguistic, contextual, social and cultural experiences associated
with multilingualism (Bialystok 2017; Grosjean 2015) should promote (the development of)
FGA. Hence, multilingualism is expected to predict FGA.
Hypothesis 2: The relationship between multilingualism and FGA is mediated by EFs. If cogni-
tive (executive) functions were supporting or even necessary conditions for FGA, and if the ‘narrow’
CA hypothesis was tenable (i.e. if multilingualism enhances EFs), then the assumed relationship
between multilingualism and FGA (hypothesis 1) is expected to be partly mediated by the ‘basic’
cognitive functions addressed in the ‘narrow’CA Hypothesis (i.e. mediated through EFs).
Note that hypothesis 1 neither presupposes an effect of multilingualism on EFs nor does it
assume EFs to be supportive or constitutive for FGA (see Fan et al. 2015, for a similar argument).
Accordingly, the test of hypothesis 1 neither depends on the corroboration nor on the refutation of
hypothesis 2. However, hypothesis 2 entails hypothesis 1. Furthermore, hypothesis 2 entails a test
(replication) of the ‘narrow’CA hypothesis. We would like to emphasise that testing the CA
hypothesis in its narrow sense is not the main intention of the present study. If it were the intention
to refute the CA hypothesis, a much larger sample with greater statistical power would be necessary
(among several other conditions; see discussion). If, on the other hand, a corroboration of the CA
hypothesis would be the aim of the study, the choice of methods of assessment for executive func-
tions that were used in earlier studies supporting the (narrow) CA hypothesis might be to less severe
(beyond mere replication). Rather, the central intention of hypothesis 2 of the present study is to
test an explanation of the development of FGA (which entails a re-test of the (narrow) CA
hypothesis).
Empirical study
Participants
The sample consisted of 119 participants between 19 and 79 years of age (M= 36.4), 77 of them were
female. Data were elicited through an online study (from May to July 2019), resulting in an ‘inci-
dental’sample. The largest group of participants lives in Germany (95 participants, 79.8%), with the
United States as the second largest group (10 participants, 8.4%), followed by the United Kingdom
(6 participants, 5%). Other countries include Canada, Sweden, Belgium, Denmark, Italy and Swit-
zerland with one or two participants per country. The statistics on the country of birth are similar,
with the largest group again being born in Germany (90 participants, 75.6%), the United States as
the second largest group (6 participants, 5%), followed by the United Kingdom (5 participants,
4.2%). Other countries of birth are China, Sweden and Switzerland with 2 participants (1.7%)
each, followed by Belgium, Canada, Denmark, India, Italy, Kazakhstan, Mexico, the Netherlands,
Russia, Slovakia, Vietnam and Tunisia with 1 participant (0.8%) each.
Method
Questionnaire
Participants were asked to give information about their socioeconomic status (education and occu-
pation of the participants and their parents [ISEI, Ganzeboom 2010]), their handedness, whether
their eyesight is impaired and whether they have any neurological impairments or disabilities.
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 5
These questions were important to rule out outside factors that could influence the results from the
measures of executive functions (see below).
Degree of multilingualism. Anderson et al. (2018) developed the language and social background
questionnaire (LSBQ) to measure degree of multilingualism. The questionnaire was originally
developed for young adults. For the purpose of this study and in order to keep the questionnaire
as brief and simple as possible, parts of the LSBQ were adapted to assess the participants’degree
of multilingualism: We used the questions on language background (section two of the LSBQ,
items 15–17.2) and slightly adapted the wording of the questions. Explanations were added as
the questionnaire was carried out online without an interviewer to help clarify items. The possibility
to add more than two languages was given to accommodate for multilingual speakers. The ques-
tionnaire was then translated into German, and participants were asked to choose between the Eng-
lish and German version of the questionnaire. Many participants reported speaking more than two
languages, but their proficiency in these languages varied greatly. It was therefore decided to only
include the languages with the highest proficiency and usage ratings and to calculate the value for
the (up to) three best languages. The resulting indicator for degree of multilingualism used in this
study thus comprises the mean of the respective self-ratings for speaking, understanding, reading,
and writing for each of the (up to) three best languages (‘Relative to a native speaker (who learned
the language from birth), how would you rate your proficiency level on a scale of 0–10 for the fol-
lowing activities in this particular language: speaking, understanding, reading, writing?’) and of
intensity of usage (‘Please indicate how much time on average you spend each week on this skill
in this particular language for speaking, listening, reading, and writing (none –little –some –
most –all)’). Some participants did not rank the languages according to their (self-assessed)
fluency and competence. In those cases, all ratings (including L4 or L5, if applicable) were used
to determine the three best languages, if necessary.
Multilingual competence groups. Participants were classified as (mainly) monolingual if they did
not indicate another language, or if their ratings of an L2 were low (i.e. less than 5 on a 10-point
scale). In order to be able to differentiate within the multilinguals between persons with (relatively)
early and (somewhat) later language acquisition, persons whose multilingual language acquisition
started before the age of three (early childhood multilinguals; ECM) were distinguished from per-
sons whose mutlingual language acquisition had started after the age of three years (late childhood
or adulthood multilinguals; LCAM). For the purposes of this study, it seemed less important to dis-
tinguish between simultaneous and sequential second language acquisition (boundaries would also
be arbitrary here and are used differently), but rather to address in particular the early developmen-
tal stimulation of executive functions (before they can be differentially diagnozed, which is the case
from about the third year of life onwards; for comprehensive overviews see Bjorklund and Causey
2018, 260ff; Goswami 2008, 295ff). The chosen age limit follows the reasoning that the influence of
the demands associated with multilingual language acquisition should be particularly strong at
early, formative cognitive developmental stages (in addition, the earlier the multilingualism begins,
the longer its (presumed) effects can ‘operate’). According to this differentiation, the sample con-
tained three multilingual competence groups (MLC groups): 35 (mainly) monolinguals (MM), 17
early childhood multilinguals (ECM; age of acquisition before age 3) and 67 late childhood or adult-
hood multilinguals (LCAM; age of acquisition after age 3).
As Table 1 shows, (self-reported) language proficiency for L2 and L3 is not generally lower for
relatively late multilinguals; this might indicate that the incidental sampling was differentially selec-
tive for these MLC groups. In addition, the mean age of relatively early multilinguals is significantly
lower than the mean age of monolinguals (F= 3.86, p= .024; monolinguals and late multilinguals do
not differ significantly). Table 2 gives more detailed information about the number of monolinguals
and multilinguals in three different age groups.
Based on information concerning occupation, socioeconomic status (SES) was measured using
ISEI (Ganzeboom 2010), which was averaged across participants and their parents (where scores
were available). Although SES is somewhat lower for early multilinguals than for late multilinguals
6W. GREVE ET AL.
and for monolinguals, the differences with respect to the SES of the participants and their
families between the three language groups are not significant (F= 2.51; p= .086). Separate ANO-
VAs for the ISEI of the participants and their parents showed no significant differences between
MLC groups as well.
Flexible Goal Adjustment. FGA was measured using the scale from Brandtstädter and Renner
(1990). This scale consists of 15 items. Sample items for the FGA scale are: „In general, I am not
upset very long about an opportunity passed up; ‘I can adapt quite easily to changes in a situation;’
‘After a serious disappointment, I soon turn to new tasks;’‘Even if everything goes wrong, I still can
find something positive about the situation’.
2
The internal consistency of the FGA scale was α= .86.
It is worth mentioning that, contrary to the majority of earlier studies (e.g. Brandtstädter and Greve
1994; Brandtstädter and Rothermund 2002), we found no bivariate correlation for age and FGA
(possibly as a consequence of a selective sample).
Assessment of executive functions (EFs)
In the present study, two assessments for executive functions were administered to operationalise
executive functions, a Stroop task and a flanker task. Both of these tasks were used in several pre-
vious studies to investigate the CA hypothesis (flanker task: Kousaie and Phillips 2017; Paap and
Greenberg 2013; Poarch and Bialystok 2015; Emmorey et al. 2008; Costa et al. 2009; Stroop task:
Bialystok and DePape, 2009; Blumenfeld and Marian 2014; Duñabeitia et al. 2014; Kousaie and
Phillips 2017; for overviews see Antoniou 2019; Bialystok 2017; Valian 2015; Van den Noort
et al. 2019; see also Poarch and Krott 2019). For the present study, it is of particular importance
that both tasks were used in a similar form in studies that had provided supporting evidence for
the CA hypothesis.
Flanker. The flanker task was adapted from a study from Luk, de Sa, and Bialystok (2011). In
their study, participants were presented with four black and one red chevron and were asked to
indicate the direction the red chevron was pointing to by pressing a mouse button on either side
of the screen. In the control condition, only the red chevron appeared in the middle of the screen.
In the congruent and incongruent conditions, the red chevron was always flanked by four black
chevrons. The red chevron could appear in any one of the three central positions. In congruent
trials, the chevrons all pointed in the same direction, whereas in the incongruent trials, the red chev-
ron would point in the direction opposite of the flanking chevrons. The experiment consisted of two
Table 1. Means and standard deviations for participants’age and language proficiency (self-reports) for multilingual competence
groups (MLC groups).
MLC group MM ECM LCAM
Age (years) 41.72 (16.56) 29.47 (10.83) 34.60 (15.48)
age of acquisition L2 10.13 (1.73) 1.67 (3.20) 9.57 (3.71)
age of acquisition L3 11.17 (3.92) 8.29 (5.25) 14.94 (7.10)
L1 7.01 (2.19) 8.49 (1.24) 8.83 (1.29)
L2 3.55 (1.34) 6.38 (1.82) 7.01 (1.71)
L3 3.44 (.09) 5.63 (2.26) 4.65 (1.99)
use of 3 ‘best’languages 8.89 (3.53) 20.33 (3.49) 18.66 (3.82)
Note: MM: (mainly) monolinguals; ECM: early childhood multilinguals (age of acquisition < 3 yrs.), LCAM: late childhood or adult-
hood multilinguals (age of acqusition > 3 yrs.)
Table 2. Age of participants for MLC-groups.
MLC group N MM ECM LCAM
19–30 55 12 10 33
31–45 22 7 4 11
46–79 28 13 1 14
Note: MM: (mainly) monolinguals; ECM: early childhood multilinguals (age of acquisition < 3 yrs.), LCAM: late childhood or adult-
hood multilinguals (age of acqusition > 3 yrs.)
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 7
control blocks with 12 trials in each block, and two conflict blocks with 24 congruent and 24 incon-
gruent trials per block. The order of the trials in each block was randomised and they were counter-
balanced for direction of the target arrow and trial type. In the present study, the procedure
corresponds to the one used by Luk, de Sa, and Bialystok (2011), but instead of answering by press-
ing a mouse button, participants indicated the direction of the red chevron by pressing a key on the
left (‘A’) or right side (‘K’) of the keyboard. The stimulus was presented with a maximum of 2000 ms
or until a response was detected, and with a 500 ms break in between trials. The blocks appeared in
the order ‘control –conflict –control –conflict’with a break in between the first conflict and the
second control block. The participants had to press a key when they wanted to continue with the
second control block. This was unavoidable due to restrictions on the length and amount of trials
that could be programmed into one question in the online survey programme. The reaction times
and accuracy were measured for each trial, and only trials where a correct response was given were
analysed. Across all trials, the reliability of the flanker task is rather high (Cronbach’sα= .971, split
half (Guttman) = .935).
Stroop. Since the publication of Stroop’s original article in 1935, the Stroop Task has been
used by many researchers in many variations (MacLeod 1991). Some researchers still use the
original Stroop Color Word Test, but a single stimulus version is more common in recent
research, in which only one colour word is presented at a time. For this paper, a single-stimulus
version with a manual response was chosen, as participants were completing the task on their
own computers. The Stroop task was adapted from Bialystok, Craik, and Luk (2008). In this
study, the stimuli consisted of the words ‘red’,‘green’and ‘blue’. They were presented in the
centre of the screen, and participants responded orally. The experiment consisted of four con-
ditions, and their order was counterbalanced across participants. In the control condition, an
array of four Xs in one of the target colours was presented to account for colour-naming
speed, and in the word reading condition, participants were asked to read the target colour
words presented in black font. In the congruent condition, the colour words presented corre-
sponded with the font they were presented in, and in the Stroop condition, the colour words
presented were conflicting with the font they were presented in. There were two blocks for
each condition, with 24 four trials in each block. For the present study, the task was changed
slightly in order to collect the responses within an online survey. Participants had to respond
manually by pressing a key on the keyboard. Each colour was assigned to a key, with the key
‘X’for the colour blue, ‘V’for the colour green and ‘L’for the colour red. Furthermore, the
word reading speed condition was left out since it could not be controlled for in this setting
as there was no way to control whether participants read the words aloud; measurement
would have been inaccurate as participants would have to read the word and press a key sim-
ultaneously. Therefore, the adapted experiment consisted of 6 blocks with 24 trials in each block.
The order the blocks appeared in was always the same (colour control –congruent –Stroop –
congruent –Stroop –colour control) in contrast to the random order of blocks in the original
trial. Similar to the flanker task, after the first three blocks participants had to press a button to
start with the last three blocks due to constraints in the design of questions in the web appli-
cation. Across all trials, the reliability of the Stroop task is rather high (Cronbach’sα= .997, split
half (Guttman) = .999).
The assessment for both indicators for EFs was conducted using the same online platform used
to collect self-report data (SoSci Survey; Leiner 2020; version 3.2.16; www.soscisurvey.de). Socisur-
vey states (https://www.soscisurvey.de/help/doku.php/de:create:questions:assignment) that reac-
tion times are recorded with an accuracy of 10 ms (JavaScript based).
For both tasks, participants were excluded if they produced too many errors (i.e. 3rd quartile +
1.5 * interquartile-difference). Additionally, for each participant reaction times were excluded if
either the response was wrong or if the reaction time were too long (i.e. 3rd quartile + 1.5 * inter-
quartile-difference). No significant differences between the MLC-groups regarding EFs were found
in an ANOVA (F(2,102) = 0.76, p= .472). (Table 3).
8W. GREVE ET AL.
Results
To test hypothesis 1, in a first step the bivariate relationship between language competence and FGA
was calculated. For degree of multilingualism, this results in a bivariate correlation of r= .28, p
= .004 (see Figure 1).
In addition to this individual correlation, an analysis of variance (oneway) for the MLC groups
yielded a significant effect (F(2, 102) = 81.77; p< .001; see Table 4 for means and SDs); the post-hoc
analysis shows that the FGA scores of monolinguals differ significantly from both multilingual
groups (p< .001), whereas the two multilingual groups do not differ from each other.
This confirms and enhances the result of the bivariate correlation: The significant contrast
between monolinguals and multilinguals particularly supports the assumption that multilingualism
can be seen as (a proxy for) a promoting condition for (the development of) FGA.
To test hypothesis 2, several mediator analyses were calculated. To that end, in a first step, both
the flanker and the Stroop task were calculated as single indicators of the EFs, respectively. In a
second step, we calculated the (standardised) mean of both tasks as a common EF indicator in
order to overcome possible weakness of each of these two measures (actually, both tasks assess sep-
arate aspects of EFs, as a weak correlation indicates: r= .15, p= .146). The bivariate examination
shows that neither the relationship between degree of multilingualism and EF (averaged over
both indicators) becomes significant (r=−.08, p= .396 (power = .55, as calculated with G*Power;
Faul et al. 2007)) nor the bivariate relationship between EF and FGA (r=−.04, p= .656 (power
= 0.69)). Accordingly, we found no mediating effect of cognitive control on the relation between
degree of multilingualism and FGA, as the simple mediation model analysis proved to be not sig-
nificant. The mediator path coefficients were not significant (a=−1.92, p= .396; b<−.001, p
= .827). Accordingly, the indirect effect was not significant (ab = .00, 95% CI =−.0009, .0040). How-
ever, the direct effect (multilingualism to FGA) was significant (c’= .03, p= .005), corroborating
hypothesis (1) controlling for EFs. Very similar results occur if the flanker and the Stroop indicators
Table 3. EF means and SDs for MLC-groups.
MM ECM LCAM
M 358.05 305.63 345.58
SD 143.80 66.22 146.37
Note: MM: (mainly) monolinguals; ECM: early childhood multilinguals (age of acquisition < 3 yrs.), LCAM: late childhood or adult-
hood multilinguals (age of acqusition > 3 yrs.)
Figure 1. Scatterplot with individual results and regression from FGA on language proficiency (degree of multilingualism).
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 9
were used as EF indicators separately (flanker: a=−1.74 (p= .399), b= .0005 (p= .277), indirect
effect ab =−.0009 (LL: −.0037; UL: .0013), c’= .03 (p= .003); Stroop: a=−2.25 (p= .177), b=
−.001 (p= .067), indirect effect ab = .002 (LL: −.0010; UL: .0095), c’= .02 (p= .012)).
Although this rejects hypothesis 2, we analysed exploratively whether the mediator hypothesis
could be demonstrated considering age as an influential factor, with regard to two aspects. If the
assumption is correct that especially early-onset multilingualism during the cognitively formative
years is influential for the development of EFs, then it would be conceivable that the effect of multi-
lingualism on EFs –and thus the first mediation pathway –occurs especially or only for relatively
early acquired multilingualism. To check whether age of acquisition has a moderating effect on the
CA hypothesis, we added MLC group as a moderator to the mediator analysis, i.e. as a moderator to
the path from multilingualism to the integrated EF indicator. However, this extended model also
showed no significant moderated direct effects (p= .997), and the index of moderated mediation
also did not reach significance (index = −.0001, 95% CI =−.0014, .0019). This suggests that the
age of acquisition of the second language (with a cut-offaround the age of three years) does not
seem to have an influence on the relationship between multilingualism and EFs in this study.
Second, the testing of the mediator hypothesis might be influenced by the fact that EFs are sub-
ject to development beyond childhood (i.e. decrease with age). Actually, several authors (e.g. Anto-
niou 2019; Bialystok et al. 2009) suggest that age might influence CA effects (studies indicate
stronger CA effects in (late) adulthood, whereas effects seem to be weaker for adolescent partici-
pants). Actually, EF (mean) showed a significant correlation with participants’age (r= .55, p
< .001). To check whether age of participants has a moderating effect on the CA hypothesis, we
added the participants’age as a moderator to the mediator analysis, i.e. as a moderator to the
path from multilingualism to the integrated EF indicator and as a moderator to the path from
EFs on FGA. A moderated mediation showed that age does not moderate the effect of multilingu-
alism on EFs (b= .00, p= .251) nor moderates the effect of EFs on FGA (b= .01, p= .059). However,
a positive main effect of age on FGA was found (b= .01, p= .005), as claimed by the two process
model (Brandtstädter and Rothermund 2002). In this model, multilingualism again shows a direct
effect on FGA (b= .03, p= .003), confirming hypothesis 1 even if controlled for age of participants.
Discussion
The present study aims to extend the CA hypothesis with regard to a broader perspective on more
general cognitive abilities, in particular to FGA. FGA is conceptualised as the individual’s ability to
dissolve blocking of goals or life problems through various cognitive adaptations such as reinterpre-
tations or selective comparisons. It involves, thus, the ability to change, vary, and integrate perspec-
tives. In particular, we hypothesised that multilingualism may enhance FGA either as an indicator
for more heterogeneous developmental conditions (hypothesis 1) or via a promotion of executive
functions (hypothesis 2). With respect to hypothesis 1, the predicted correlation between multilin-
gualism and FGA was confirmed although monolinguals are somewhat older in the present sample,
which should work against this relationship since older individuals show, in general, higher FGA
scores (Brandtstädter 2006; Brandtstädter and Rothermund 2002; we did not find a significant cor-
relation with age in the present sample; however in the moderated mediation analysis with respect
to hypothesis 2, an age affect proved significant). These findings support the suggestion to extend
Table 4. FGA means and SDs for MCL groups.
MM ECM LCAM
M 9.08 21.58 18.97
SD 3.78 3.52 4.07
Note: MM: (mainly) monolinguals; ECM: early childhood multilinguals (age of acquisition < 3 yrs.), LCAM: late childhood or adult-
hood multilinguals (age of acqusition > 3 yrs.)
10 W. GREVE ET AL.
the CA hypothesis from ‘narrow’cognitive effects (e.g. on EFs) to a broader view on cognitive flexi-
bility promoted by multilingualism (Fan et al. 2015).
With respect to hypothesis 2, the present study was not able to confirm a mediating role of EFs
on the relationship between multilingualism and FGA, although it examined two different indi-
cators for EFs. If these analyses were controlled for age of participants (which was substantially cor-
related with the EFs: r= .55) we found that the effect of EF on FGA proved significant. These effects
of EFs could indeed be an indication that the recording of EFs (despite a certain imprecision due to
the survey instrument) actually had a sufficient degree of validity. This, in turn, adds a bit of weight
to the possible interpretation that the lacking mediator effect (hypothesis 2) might be due to the lack
of an effect from multilingualism to EFs (i.e. the non-significant CA, as it were). (In additional
regression analyses, for which the multilinguals are grouped together and tested against the mono-
loguals as potential predictors of EFs, neither an effect for multilingualism –even when controlling
for participant age –nor a moderation of this path by age could be demonstrated.)
In fact, finding a corroboration of hypothesis 1 and a refutation of hypothesis 2 is consistent with
the results from Fan et al. (2015) with preschool children, indicating that multilingualism enhances
cognitive flexibility (in a broader sense) without an improving effect of multilingualism on EFs (i.e.
without presuming the narrow CA hypothesis). Although the failed replication of the (narrow) CA
was conducted using EF assessements (flanker and Stroop tasks) which have been used in previous
studies that support the CA hypothesis (Bialystok, Craik, and Luk 2008; Luk, de Sa, and Bialystok
2011), lack of evidence must not be misinterpreted as evidence for lack (in particular, since the
power for the analysis of the relationship between degree of multilingualism and EF is
rather low in our study). However, it is fair to say that this result certainly does not add any
evidence in favour of the narrow CA hypothesis, although the procedures obtained were selected
in favour of it.
Limitations
Several caveats should be noted with respect to the interpretation of the present findings. First, the
present study was conducted as an online study, resulting in an ‘incidental’, self-selective sample, an
hence in unbalanced distributions with respect to socio-cultural variables. In particular, the unequal
group sizes with respect to language proficiency implies a limitation of the robustness of the
findings with respect to group comparisons (less so for correlationel/regression analyses). This cer-
tainly underscores the need for replication (and extension) of this study. Moreover, as in many
other studies, in the present sample multilingualism is most probably confounded with social
and/or cultural conditions (e.g. Bialystok 2017; Bialystok and Barac 2013). Although we did not
find significant differences between the multilingual competence groups with respect to their socio-
economic status, it cannot be excluded that social or cultural heterogeneity beyond mere linguistic
heterogeneity could be an important developmental condition for the individual FGA and could
therefore (partially) explain the results confirming hypothesis 1. Unfortunately, the present sample
is too small to allow a systematic matching according to important indicators of SES and other
socio-cultural conditions (Czapka et al. 2020). As a consequence, the present study cannot disen-
tangle which of these possible factors are causally relevant for the promotion of FGA. Thus, a larger
and more balanced or stratified sample is certainly an important requirement for future studies.
Second, with respect to the (narrow) CA hypothesis, the online approach of the present study
implies several adaptations in the assessment procedure of the EFs: The procedure for both the
flanker and the Stroop task deviate in some details from the classical procedure (for a critical dis-
cussion of the lack of methodological rigour in this respect see Zhou and Krott 2016; for a recent
example of an online study with a very large sample and high methodical rigour see Nichols et al.
2020). We chose online data collection mainly for practical reasons: A comparable sample would
not have been available regionally in direct contact. Moreover, the assessment of the EFs with a
module downloadable to the respective local computer (such as psychopy) would in all probability
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 11
have resulted in an additional selection (mistrust regarding data security or anonymity). The
reduction in the reliability of the measures of central executive functions (which was, to be sure,
not detectable with respect to the internal consistency of both indicators) accepted with this pro-
cedure is presumably unsystematic –there is nothing to suggest that it systematically biases in
favour or to the disadvantage of the CA hypothesis (actually, in one of the moderated mediator ana-
lyses, we found, as expected, an effect from EF to FGA). However, the reduction of reliability with
respect to the EF assessments makes testing of the (narrow) CA hypothesis more conservative: If the
effect could have been shown with this procedure, this would have been strong evidence in favour of
the CA hypothesis, but the absence of the effect is not a strong argument against it. One can read
this constellation in a way that the less reliable recording of the central executive functions increases
the beta error of the interpretation of the null effect, but it is also true to say that (in this sample) the
CA effect could not have been strong (enough).
Third, the assessment of (the degree of) multilingualism leaves some room for improvement as
well. In particular, it is certainly appropriate to distinguish not only degrees but also various forms
(e.g. performances, contexts, type and frequency of switching) of multilingualism (Antoniou 2019).
For instance, it seems plausible to assume that the linguistic (dis)similarity of the respective
languages a multilingual individual uses might influence both the narrow and the broad CA (see
Antoniou and Wright 2017). With respect to the test of hypothesis 2 (i.e. CA hypothesis narrow)
in particular, it would be certainly necessary to assess the individuals’language (and cultural) profi-
ciency by objective test or indicators.
As a consequence, both the assessment of degree of multilingualism and of the EF’s capacities are
to a certain degree vulnerable to methodological objections. We hold, however, that this objection
does not render the confirmation of hypothesis 1 invalid –even if its interpretation remains to be
clarified.
A particular reason to report the results with respect to hypothesis 2, despite possible objections
against the reliability of the online assessment of the EFs, was that in order to at least partially com-
pensate for the reliability reduction of the EF assessment to the disadvantage of the CA hypothesis,
we adopted the procedure from two available confirmatory studies of some of the prominent pro-
ponents of the CA hypothesis (Bialystok, Craik, and Luk 2008; Luk, de Sa, and Bialystok 2011). This
procedure should have given the CA hypothesis an overall fair chance of being corroborated,
despite of some error-related ‘noise’in the measurement of the online data collection. In order
to reduce the publication bias discussed in relation to the CA hypothesis (de Bruin, Treccani,
and Della Sala 2015), it seemed right and important to us not to omit it.
Heterogeneity as a promoting factor for flexibility of goal adjustment
The finding of a prediction of FGA by multilingualism both individually (correlation) and by MLC
group (ANOVA) converges with earlier findings that heterogeneity (mediated by divergent think-
ing) is a promoting condition for FGA (Greve and Thomsen 2013,2016; Thomsen and Greve 2013).
Due to the confounding of linguistic, social and cultural heterogeneity in the variable multilingu-
alism (Grosjean 2015), the factors that are actually effective cannot be examined more closely here.
Therefore, in addition to the necessity of a replication of these findings and improvement of the
measures both for multilingualism and EFs, a further demand for future research is the systematic
separation of possible stimulating developmental factors through systematic sampling with respect
to cultural and social conditions.
It is important to note that we did not find a significant correlation for age with FGA, contrary to
what many other studies have reported (Brandtstädter 2006; Brandtstädter and Rothermund 2002).
Although we actually found an age effect for one of the moderated mediation analyses (see above),
this underscores the necessity of a replication of the finding of the present study with a larger and
more representative sample. Additionally, such a replication study should include an indicator for
divergent thinking as a possible mediator of the relationship between multilingualism and FGA.
12 W. GREVE ET AL.
Moreover, the investigation of the role of certain cognitive capacities for FGA is an important
requirement for future studies. In this respect, particularly the use of prospective longitudinal
studies is essential for tests of causal hypotheses (e.g. cross-lagged panel); this applies to the inves-
tigation of the developmental conditions of FGA as well as the developmental consequences of
multilingualism.
Overcoming the (too) narrow cognitive advantage hypothesis
Hence, the present study, with the limitations discussed above, must not be seen as conclusive
evidence against the (narrow) CA hypothesis. A severe test of it would certainly need a systema-
tic study with a large sample (i.e. sufficient power) of various types of multilinguals and mono-
linguals, a comprehensive test battery (including Stroop and Flanker tasks, among several
others), and, in particular, valid and objective language tasks. However, given that the present
sample includes a majority of adults (which, as a rule, have produced more CA-confirming
results: Antoniou 2019), and given that we selected EF indicators from confirming studies,
our results certainly do not add support for the narrow CA hypothesis. In view of recent results
and current meta-studies (Lehtonen et al. 2018; Nichols et al. 2020; Paap, Johnson, and Sawi
2015; as well as the methodological criticism of CA-supporting meta-analyses: Paap et al.
2020), the possibility that the narrow CA hypothesis might depend on several conditions (i.e.
moderators; Antoniou and Wright 2017; Festman, Czapka, and Winsler 2021) should be
seriously considered. In addition, beyond the ubiquitous criticism with respect to the methodical
rigour in the application of Stroop and flanker tasks (Zhou and Krott 2016), there are numerous
indications that allegedly EF-assessing measures such as these are reliably replicable as a
phenomenon at the group level, but are not sufficiently reliable as individual diagnostics
(Hedge, Powell, and Sumner 2018). This might also explain, at least partially, the incoherent
findings with respect to the relationship between these EF indicators and multilingualism,
even beyond the much-discussed influence of moderating factors (Paap, Johnson, and Sawi
2015; Lehtonen et al. 2018).
Irrespective of these considerations, the present study can be seen as supporting a case for the
extension of the (narrow) CA hypothesis. Irrespective of the lacking EF-mediator effect in this
study, we found a cognitive advantage for multilinguals with respect to one prototypical facet of
cognitive flexibility in a broader sense: flexibility of goal adjustment. In addition to recent proposals
to extend the CA debate into a neuroscientific direction (Antoniou 2019) or with respect to the indi-
vidual’s self-concept (Festman and Schwieter 2019), and in line with previous studies concerning an
advantage of multilinguals with respect to (cognitive) flexibility (Dewaele and Botes 2019; Fan et al.
2015), we propose to take into account other and broader cognitive advantages that may be pro-
moted by multilingualism. The results of our study suggest that the inclusion of indicators for cog-
nitive flexibility (e.g. divergent thinking) as possible mediators in a replication study could be
interesting as a possible extension of the CA hypothesis beyond the explanation of the developmen-
tal conditions of FGA.
Notes
1. Terminology concerning the terms bilingualism, tringualism and multilingualism is not consistent across
publications. While some authors use the term bilingualism as an umbrella term for all types of
additional languages (L2, L3, …Ln, e.g., Ellis 2003), other authors differentiate between these terms
(e.g., Festman 2019). In this paper, we use the term multilingualism to include the knowledge of
more than one language (two, three, or more languages) independent of the time of acquisition (at
birth or later in life).
2. The entire questionnaire from Brandtstädter and Renner (1990) comprises two scales: In addition to the FGA
scale, the scale for tenacious goal pursuit (TGP) consists of 15 items as well. This scale was not analysed in the
present study.
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 13
Disclosure statement
No potential conflict of interest was reported by the author(s).
References
Adesope, O. O., T. Lavin, T. Thompson, and C. Ungerleider. 2010.“A Systematic Review and Meta-Analysis of the
Cognitive Correlates of Bilingualism.”Review of Educational Research 80: 207–245. doi:10.3102/
0034654310368803.
Anderson, J. A. E., L. Mak, A. Keyvani Chahi, and E. Bialystok. 2018.“The Language and Social Background
Questionnaire: Assessing Degree of Bilingualism in a Diverse Population.”Behavior Research Methods 50 (1):
250–263. doi:10.3758/s13428-017-0867-9.
Antoniou, M. 2019.“The Advantages of Bilingualism Debate.”Annual Review of Linguistics 5: 395–415.
Antoniou, M., and S. M. Wright. 2017.“Uncovering the Mechanisms Responsible for Why Language Learning May
Promote Healthy Cognitive Aging.”Frontiers in Psychology 8: 2217. doi:10.3389/fpsyg.2017.02217.
Baker, C., and W. E. Wright. 2017.Foundations of Bilingual Education and Bilingualism. 6th ed. Bristol: Multilingual
Matters.
Bialystok, E. 2015.“Bilingualism and the Development of Executive Function: The Role of Attention.”Child
Development Perspectives 9: 117–121. doi:10.1111/cdep.12116.
Bialystok, E. 2017.“The Bilingual Adaptation: How Minds Accommodate Experience.”Psychological Bulletin 143 (3):
233–262. doi:10.1037/bul0000099.
Bialystok, E., and R. Barac. 2013.“Cognitive Effects.”In The Psycholinguistics of Bilingualism, edited by F. Grosjean,
and P. Li, 192–213. Chichester, UK: Wiley-Blackwell.
Bialystok, E., F. I. M. Craik, D. W. Green, and T. H. Gollan. 2009.“Bilingual Minds.”Psychological Science 10 (3):
89–129.
Bialystok, E., F. I. Craik, and G. Luk. 2008.“Cognitive Control and Lexical Access in Younger and Older Bilinguals.”
Journal of Experimental Psychology: Learning, Memory, and Cognition 34 (4): 859–873. doi:10.1037/0278-7393.34.
4.859.
Bialystok, E., F. I. Craik, and G. Luk. 2012.“Bilingualism: Consequences for Mind and Brain.”Trends in Cognitive
Sciences 16 (4): 240–250. doi:10.1016/j.tics.2012.03.001.
Bialystok, E., and A. M. DePape. 2009.“Musical Expertise, Bilingualism, and Executive Functioning.”Journal of
Experimental Psychology: Human Perception and Performance 35: 565–574.
Bialystok, E., and M. M. Martin. 2004.“Attention and Inhibition in Bilingual Children: Evidence from the
Dimensional Change Card Sort Task.”Developmental Science 7: 325–339.
Bialystok, E., and M. Viswanathan. 2009.“Components of Executive Control with Advantages for Bilingual Children
in Two Cultures.”Cognition 112: 494–500.
Bjorklund, D. F., and K. B. Causey. 2018.Children’s Thinking. Cognitive Development and Individual Differences. Los
Angeles: Sage.
Blanco-Elorrieta, E., and L. Pylkkänen. 2018.“Ecological Validity in Bilingualism Research and the Bilingual
Advantage.”Trends in Cognitive Sciences 22: 1117–1126.
Blumenfeld, H. K., and V. Marian. 2014.“Cognitive Control in Bilinguals: Advantages in Stimulus-Stimulus
Inhibition.”Bilingualism 17: 610–629.
Brandtstädter, J. 2006.“Action Perspectives on Human Development.”In Handbook of Child Psychology: Vol.1.
Theorectical Models of Human Development, edited by W. Damon, and R. M. Lerner, 6th ed., 516–568.
New York: Wiley.
Brandtstädter, J. 2007.Das flexible Selbst. München: Elsevier.
Brandtstädter, J. 2015.Positive Entwicklung. Zur Psychologie gelingenden Lebens. Berlin: Springer.
Brandtstädter, J., and W. Greve. 1994.“The Aging Self: Stabilizing and Protective Processes.”Developmental Review
14: 52–80.
Brandtstädter, J., and G. Renner. 1990.“Tenacious Goal Pursuit and Flexible Goal Adjustment: Explication and Age-
Related Analysis of Assimilative and Accommodative Strategies of Coping.”Psychology and Aging 5: 58–67.
Brandtstädter, J., and K. Rothermund. 2002.“The Life Course Dynamics of Goal Pursuit and Goal Adjustment: A
Two-Process Framework.”Developmental Review 22: 117–150.
Costa, A., M. Hernández, J. Costa-Faidella, and N. Sebastián Gallés. 2009.“On the Bilingual Advantage in Conflict
Processing: Now You See It, Now You Don’t.”Cognition 113: 135–149.
Czapka, S., A. Klassert, and J. Festman. 2019.“Executive Functions and Language: Their Differential Influence
on Mono- vs. Multilingual Spelling in Primary School.”Frontiers in Psychology 10: 97. doi:10.3389/fpsyg.2019.
00097.
Czapka, S., C. Wotschack, A. Klassert, and J. Festman. 2020.“A Path to the Bilingual Advantage: Pairwise Matching
of Individuals.”Bilingualism: Language and Cognition 23: 344–354. doi:10.1017/ S1366728919000166.
14 W. GREVE ET AL.
de Bruin, A., B. Treccani, and S. Della Sala. 2015.“Cognitive Advantage in Bilingualism an Example of Publication
Bias?”Psychological Science 26 (1): 99–107.
De Groot, A. M. B. 2011.Language and Cognition in Bilinguals and Multilinguals. New York: Psychology Press.
Dewaele, J.-M., and E. Botes. 2019.“Does Multilingualism Shape Personality? An Exploratory Study.”International
Journal of Bilingualism 23: 1–13.
Duñabeitia, J. A., J. A. Hernández, E. Antón, P. Macizo, A. Estévez, L. J. Fuentes, and M. Carreiras. 2014.“The
Inhibitory Advantage in Bilingual Children Revisited: Myth or Reality?”Experimental Psychology 61: 234–251.
Ellis, R. 2003.Second Language Acquisition. Oxford: Oxford University Press.
Emmorey, K., G. Luk, J. E. Pyers, and E. Bialystok. 2008.“The Source of Enhanced Cognitive Control in Bilinguals:
Evidence from Bimodal Bilinguals.”Psychological Science 19: 1201–1206.
Fan, S. P., Z. Liberman, B. Keysar, and K. D. Kinzler. 2015.“The Exposure Advantage: Early Exposure to a
Multilingual Environment Promotes Effective Communication.”Psychological Science 26 (7): 1090–1097.
Faul, F., E. Erdfelder, A.-G. Lang, and A. Buchner. 2007.“G*Power 3: A Flexible Statistical Power Analysis Program
for the Social, Behavioral, and Biomedical Sciences.”Behavior Research Methods 39: 175–191.
Festman, J. 2019.“Lecture 9: The Psycholinguistics of Multilingualism.”In Twelve Lectures on Multilingualism, edited
by D. Singleton, and L. Aronin, 223–269. Bristol: Multilingual Matters.
Festman, J., S. Czapka, and A. Winsler. 2021.“How Many Moderators Does Is Take Till We Know …That Too Many
Bilingual Advantage Effects Have Died?”In Understanding Variability in Second Language Acquisition,
Bilingualism, and Cognition: A Multi-Layered Perspective, edited by K. Kersten, and A. Winsler. London:
Routledge.
Festman, J., and K. Kersten. 2010.“Kognitive Auswirkungen von Zweisprachigkeit.”In CLIL und Immersion:
Erfolgsbedingungen für fremdsprachlichen Sachfachunterricht in der Grundschule, edited by U. Massler, and P.
Burmeister, 38–52. Braunschweig: Westermann.
Festman, J., and J. W. Schwieter. 2019.“Self-Concepts in Reading and Spelling among Mono- and Multilingual
Children: Extending the Bilingual Advantage.”Behavioral Sciences 9: 39. doi:10.3390/bs9040039.
Ganzeboom, H. B. 2010. International standard classification of occupations ISCO-08 with ISEI-08 scores. Version
July, 27.
Goswami, U. 2008.Cognitive Development. The Learning Brain. Hove, UK: Psychology Press.
Greve, W., B. Leipold, and C. Kappes. 2017.“Fear of Crime in Old Age: A Sample Case of Resilience?”Journals of
Geronotology Series B: Psychological Sciences and Social Sciences. doi:10.1093/geronb/gbw169.
Greve, W., and T. Thomsen. 2013.“Developmental Conditions of Adaptive Self-Stabilization in Adolescence: An
Exploratory Study.”International Journal of Developmental Science 7 (3-4): 119–131. doi:10.3233/DEV-120101.
Greve, W., and T. Thomsen. 2016.“Evolutionary Advantages of Free Play During Childhood.”Evolutionary
Psychology 14 (4). doi:10.1177/1474704916675349.
Greve, W., T. Thomsen, and C. Dehio. 2014.“Does Playing Pay? The Fitness-Effect of Free Play During Childhood.”
Evolutionary Psychology 12 (2): 434–447. doi:10.1177/147470491401200210.
Grosjean, F. 2015.“Bicultural Bilinguals.”International Journal of Bilingualism 19 (5): 572–586.
Grundy, J. G., and K. Timmer. 2017.“Bilingualism and Working Memory Capacity: A Comprehensive Meta-
Analysis.”Second Language Research 33 (3): 325–340.
Haase, C. M., J. Heckhausen, and C. Wrosch. 2013.“Developmental Regulation Across the Life Span: Toward a New
Synthesis.”Developmental Psychology 49 (5): 964–972. doi:10.1037/a0029231.
Heckhausen, J., C. Wrosch, and R. Schulz. 2010.“A Motivational Theory of Lifespan Development.”Psychological
Review 117: 32–60.
Heckhausen, J., C. Wrosch, and R. Schulz. 2019.“Agency and Motivation in Adulthood and Old Age.”Annual Review
of Psychology 70: 191–217.
Hedge, C., G. Powell, and P. Sumner. 2018.“The Reliability Paradox: Why Robust Cognitive Tasks Do Not Produce
Reliable Individual Differences.”Behavior Research Methods 50: 1166–1186.
Hilchey, M. D., and R. M. Klein. 2011.“Are There Bilingual Advantages on Nonlinguistic Interference Tasks?
Implications for the Plasticity of Executive Control Processes.”Psychonomic Bulletin & Review 18 (4): 625–658.
doi:10.3758/s13423-011-0116-7.
Hong, Y.-y., M. W. Morris, C.-y. Chiu, and V. Benet-Martínez. 2000.“Multicultural Minds: A Dynamic
Constructivist Approach to Culture and Cognition.”American Psychologist 55 (7): 709–720.
Kappes, C., and T. Thomsen. 2020.“Imitation of Goal Regulation Processes in Romantic Relationships.”European
Journal of Personality. doi:10.1002/per.2244.
Kharkhurin, A. V. 2007.“The Role of Cross-Linguistic and Cross-Cultural Experiences in Biliguals’Divergent
Thinking.”In Cognitive Aspects of Bilingualism, edited by I. Kecskes, and L. Albertazzi, 175–210. Dordrecht,
Netherlands: Springer.
Kharkhurin, A. V. 2009.“The Role of Bilingualism in Creative Performance on Divergent Thinking and Invented
Alien Creatures Tests.”The Journal of Creative Behavior 43: 59–71.
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 15
Kharkhurin, A. V. 2015.“Bilingualism and Creativity. An Educational Perspective.”In The Handbook of Bilingual
and Multilingual Education, edited by W. E. Wright, S. Boun, and O. García, 38–55. Hoboken, NJ: Wiley-
Blackwell.
Kousaie, S., and N. A. Phillips. 2017.“A Behavioural and Electrophysiological Investigation of the Effect of
Bilingualism on Aging and Cognitive Control.”Neuropsychologia 94: 23–35.
Kovács, ÁM. 2009.“Early Bilingualism Enhances Mechanisms of False-Belief Reasoning.”Developmental Science 12:
48–54.
Kramsch, C. 2004.“Language, Thought, and Culture.”In The Handbook of Applied Linguistics, edited by A. Davies,
and C. Elder, 235–265. London: Blackwell Publishers.
Kroll, J. F., P. E. Dussias, C. A. Bogulski, and J. R. V. Kroff.2012.“Juggling Two Languages in One Mind: What
Bilinguals Tell Us About Language Processing and Its Consequences for Cognition.”Psychology of Learning
and Motivation 56: 229–262. doi:10.1016/B978-0-12-394393-4.00007-8.
Laine, M., and M. Lehtonen. 2018.“Cognitive Consequences of Bilingualism: Where to Go from Here?”Language,
Cognition and Neuroscience 33 (9): 1205–1212. doi:10.1080/23273798.2018.1462498.
Lee, H., and K. H. Kim. 2011.“Can Speaking More Languages Enhance Your Creativity? Relationship between
Bilingualism and Creative Potential among Korean American Students with Multicultural Link.”Personality
and Individual Differences 50: 1186–1190.
Lehtonen, M., A. Soveri, A. Laine, J. Järvenpää, A. de Bruin, and J. Antfolk. 2018.“Is Bilingualism Associated with
Enhanced Executive Functioning in Adults? A Meta-Analytic Review.”Psychological Bulletin 144: 394–425.
Leiner, D. J. 2020. SoSci Survey (Version 3.2.18) [Computer software]. Available at: https://www.soscisurvey.de.
Lessing, N., C. Kappes, W. Greve, and C. Mähler. 2019.“Developmental Conditions of Accommodative Coping in
Childhood: The Role of Executive Functions.”Cognitive Development 50: 56–65.
Lessing, N., T. Thomsen, C. Mähler, and W. Greve. 2017.“Selbstregulation und flexible Zielanpassung im
Vorschulalter.”Kindheit und Entwicklung 26 (1): 19–27. doi:10.1026/0942-5403/a000212.
Luk, G., E. de Sa, and E. Bialystok. 2011.“Is There a Relation Between Age of Onset Age of Bilingualism and
Enhancement of Cognitive Control?”Bilingualism: Language and Cognition 14: 588–595.
MacLeod, C. M. 1991.“Half a Century of Research on the Stroop Effect: An Integrative Review..”Psychological
Bulletin 109 (2): 163–203. doi:10.1037/0033- 2909.109.2.163.
Meyer, T., and W. Greve. 2012.“Die Entwicklungsbedingungen der Adaptivität. Theoretische Überlegungen und
empirische Befunde zu einem Entwicklungsmodell akkommodativer Regulationskompetenz.”Zeitschrift für
Gesundheitspsychologie 20 (1): 27–38. doi:10.1026/0943-8149/a000059.
Nichols, E. S., C. J. Wild, B. Stojanoski, M. E. Battista, and A. M. Owen. 2020.“Bilingualism Affords no General
Cognitive Advantages: A Populations Study of Executive Function in 11,000 People.”Psychological Science 31:
548–567.
Paap, K. R., and Z. I. Greenberg. 2013.“There is no Coherent Evidence for a Bilingual Advantage in Executive
Processing.”Cognitive Psychology 66: 232–258.
Paap, K. R., H. A. Johnson, and O. M. Sawi. 2015.“Bilingual Advantages in Executive Function Either do not Exist or
are Restricted to Very Specific and Undetermined Circumstances.”Cortex 69: 265–278. doi:10.1016/j.cortex.2015.
04.014.
Paap, K. R., L. Mason, B. Zimiga, Y. Silva, and M. Frost. 2020.“The Alchemy of Conformation Bias Transmutes
Expectations into Biligual Advantages: A Tale of Two New Meta-Analyses.”Quarterly Journal of Experimental
Psychology 73 (8): 1290–1299.
Pavlenko, A. 2011.“Thinking and Speaking in two Languages: Overview of the Field.”In Thinking and Speaking in
Two Languages, edited by A. Pavlenko, 237–257. Bristol: Multilingual Matters.
Pavlenko, A. 2014.The Bilingual Mind and What It Tells Us About Language and Thought. Cambridge: Cambridge
University Press.
Piekny, J., T. Thomsen, K. Schuchardt, N. Lessing, W. Greve, and C. Mähler. 2017.“Kognitive Kompetenzen und
Kognitive Bewältigungsstrategien im Vor- und Grundschulalter.”Kindheit und Entwicklung 26 (1): 28–38.
doi:10.1026/0942-5403/a000214.
Poarch, G. J. 2018.“Multilingual Language Control and Executive Function: A Replication Study.”Frontiers in
Communication 3: 46. doi:10.3389/cfomm.2018.00046.
Poarch, G. J., and E. Bialystok. 2015.“Bilingualism as a Model for Multitasking.”Developmental Review 35: 113–124.
Poarch, G. J., and A. Krott. 2019.“A Bilingual Advantage? An Appeal for a Change in Perspective and
Recommendations for Future Research.”Behavioral Sciences 9: 95. doi:10.3390/bs9090095.
Poarch, G. J., and J. G. van Hell. 2012.“Executive Functions and Inhibitory Control in Multilingual Children:
Evidence from Second Language Learners, Bilinguals and Trilinguals.”Journal of Experimental Child
Psychology 113: 535–551. doi:10.1016/j.jecp.2012.06.013.
Rubio-Fernandez, P., and S. Glucksberg. 2012.“Reasoning About Other People’s Belief: Bilinguals Have an
Advantage.”Journal of Experimental Psychology: Learning, Memory, and Cognition 38: 211–217.
16 W. GREVE ET AL.
Rühs, F., W. Greve, and C. Kappes. 2017.“Coping with Criminal Victimization and Fear of Crime: The Protective
Role of Accommodative Self-Regulation.”Legal and Criminological Psychology 22 (2): 359–377. doi:10.1111/lcrp.
12106.
Thomsen, T. 2016.“Flexible Goal Adjustment from Late Childhood to Late Adolescence: Developmental Differences
and Relations to Cognitive Coping and Emotion Regulation.”International Journal of Developmental Science 10:
57–72. doi:10.3233/DEV-150167.
Thomsen, T., V. Fritz, R. Mößle, and W. Greve. 2015.“The Impact of Accommodative Coping on Well-Being and
Self-Esteem in Childhood and Adolescence: Longitudinal Findings.”International Journal of Behavioral
Development 39 (5): 467–476. doi:10.1177/0165025414551762.
Thomsen, T., and W. Greve. 2013.“Accommodative Coping in Early Adolescence: An Investigation of Possible
Developmental Components.”Journal of Adolescence 36 (5): 971–981. doi:10.1016/j.adolescence.2013.08.003.
Thomsen, T., C. Kappes, L. Schwerdt, J. Sander, and C. Poller. 2017.“Modelling Goal Adjustment in Social
Relationships: Two Experimental Studies with Children and Adults.”British Journal of Developmental
Psychology 35 (2): 267–287. doi:10.1111/bjdp.12162.
Titone, D., J. Gullifer, S. Subramaniapillai, N. Rajah, and S. Baum. 2017.“History-inspired Reflections on the
Bilingual Advantages Hypothesis.”In Growing Old with Two Languages: Effects of Bilingualism on Cognitive
Aging, edited by M. Sullivan, and E. Bialystok, 265–295. Amsterdam: John Benjamins Publishing.
Valian, V. 2015.“Bilingualism and Cognition.”Bilingualism: Language and Cognition 18: 3–24.
Van den Noort, M., E. Struys, P. Bosch, L. Jaswetz, B. Perriard, S. Yeo, P. Barisch, K. Vermeire, S.-H. Lee, and S. Lim.
2019.“Does the Bilingual Advantage in Cognitive Control Exist, and if so, What are ist Modulating Factors? A
Systematic Review.”Behavioral Sciences 9: 27. doi:10.3390/bs9030027.
Wisehart, M., M. Viswanathan, and E. Bialystok. 2016.“Flexibility in Task Switching by Monolinguals and
Bilinguals.”Bilingualism: Language and Cognition 19: 141–146. doi:10.1017/S1366728914000273.
Zhang, M. 2018.“An Overview of the Bilingual Advantage: History, Mechanisms, and Consequences.”Western
Undergraduate Psychology Journal 6: 1–9.
Zhou, B., and A. Krott. 2016.“Data Trimming Procedure Can Eliminate Bilingual Cognitive Advantage.”
Psychonomic Bulletin & Review 23: 1221–1230. doi:10.3758/s13423-015-0981-6.
JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 17