MULTILINGUALISM, SOCIAL COGNITION, AND

CREATIVITY

Li Wei
Birkbeck College, University of London
[email protected]
• 2014 A control process model of codeswitching (with David Green). Language,
Cognition and Neuroscience 29,4: 499-511
• 2014 The role of codeswitching in bilingual creativity and selective attention
(with Anatoliy Kharkhurin). International Journal of Bilingual Education and
Bilingualism. DOI: 10.1080/13670050.2014.884211
• 2013 Is multilingualism linked to a higher tolerance of ambiguity? Bilingualism:
Language and Cognition. 16.1: 231-240 (with Jean-Marc Dewaele)
• 2012 Multilingualism, empathy and multicompetence. International Journal of
Multilingualism 9.4: 352-366 (with Jean-Marc Dewaele)
LANGUAGE MODE OF THE BILINGUAL SPEECH
PRODUCTION (GROSJEAN, 1998)
IMPLICATIONS
• (Some) Bilinguals can behave as if they were monolingual by using only one of the
languages they know.

• How do you tell whether a person is bilingual or not?

A. Ask the person directly whether she is bilingual.
B. The person introduces herself as bilingual.
C. Someone else introduces her as bilingual.
D. You hear the person speaking two or more languages with other people.

Only in D, we have ‘evidence’ that the speaker is in fact bilingual.

• Code-Switching is therefore a defining feature of being bilingual.
• Yet, CS has received relatively little attention in cognitive psychology of bilingualism.

• Most of the existing work is on differentiation/separation/deactivation.

• (Some) Bilinguals can behave as if they were monolingual by using only one of the
languages they know.
• SOME, because there are different types of bilinguals. For some, separation is neither a
possibility nor a need.
• Experience - born into a multilingual family/community vs learning “additional” languages later
in life; living in a multilingual environment vs OLON/OLAT environment
• Multilingual reality: cf. bilinguals, few multilinguals separate languages all the time (Clyne and
others).
DIFFERENT TYPES OF CS
1. “It’s absolutely scandalistic.” (referring to a news story)
2. “Can you open the light?”
3. “lam:a fƏlik bjɪʒe” (referring to a cartoon character) (Khattab 2009: 152)
(Arabic. when Flick comes, with Flick pronounced with an epenthetic vowel, as in Arabic)
4. Girl 1: ¿Dónde estás?
(Spanish. Where are you?)
Girl 2: Upstairs.
Girl 1: ¿Dónde?
(Spanish. Where?)
Girl 2: En mi habitación.
(Spanish. In the bedroom.)
5. Mother: Nei sik muyt-ye a?
(Cantonese. What do you want to eat?)
Son: (1.0) Just apples.
Mother: Just /n/ just apples? Dimgai m sik yoghurt a?
(Cantonese. Why not have some yoghurt?)
6. “There was a guy, you know, que [that] he se montó [got up]. He started playing with
congas, you know, and se montó y empezóa brincar [got up and started to jump] and all that
shit.” (Winford, 2003: 105)
7. Ngaw wei solve di problem
(Cantonese. I will; that/those. I will solve that problem.)
8. Tu peux me pick-up-er?
(French: You can. Can you pick me up?) (Gardner-Chloros, 2009: 97)
9. I'm LAVing PANDELCAGEs.
(Danish. Make, pancake. I'm making pancakes.) (Petersen, 1988)
10. I have cha de/-ed chulai
(de – Chinese case marker. I have checked.)
Traditional de, simplified/de (de – Chinese case marker)
11. She asked me, ‘nei ha m ha-ppy la?’ So I told her, ‘ho m happy la’.
(Cantonese. You NEG. PART. Very PART. Are you happy or not? Very unhappy.)
12. Sho shenme ping!
(Mandarin. What. What shopping)
• Different structural configurations
• Different social motivations/purposes/contexts
• Different cognitive mechanism (?)
COGNITIVE MECHANISMS OF CODESWITCHING
• Cognitive control – differentiation / separation / deactivation
• Language switching and task switching
• Executive systems (cognitive advantages)

• Imaging research – neural networks involved in language switching

• Subject selection: bilinguals vs monolinguals; early vs late bilinguals; high proficiency

bilinguals vs low proficiency bilinguals
COGNITIVE CONSEQUENCES OF BILINGUALISM
• Cognitive advantages of bilingualism (knowing more than one
language)
• Non-verbal domains / reaction time (Simon task)
• Metalinguistic awareness
• Selective attention
• Creativity

• High-proficiency bilinguals have better cognitive/executive control functions

• Ability to separate languages (control/deactivation) is taken to be
the key
COGNITIVE CONSEQUENCES OF
CODESWITCHING
• What happens to habitual/dense Codeswitchers who
simply do not separate their “languages”?
• Poor executive control? Does CS require More or Less
control?
• Poor metalinguistic awareness? Grammaticality and
structural well-formedness.
• Poor selective attention, therefore poor creativity?
THE SOCIAL COGNITION OF HABITUAL
CODESWITCHERS
• A) Empathy
• B) Tolerance of Ambiguity

• with Jean-Marc Dewaele

EMPATHY
• Empathy - the ability ‘to tune into how someone else is feeling, or what they might be thinking’
(Baron-Cohen & Wheelwright, 2004, p. 193).
• Empathy plays a crucial role in social interactions as it allows us ‘to understand the intentions of
others, predict their behaviour, and experience an emotion triggered by their emotion’ (p. 193).
• Linguists working on CS often claim that multilinguals can collaboratively build sentences with
elements from different “languages”.
• Potential to test multilinguals’ Theory of Mind.
• Cognitive empathy - ‘‘the intellectual/imaginative apprehension of another’s mental state’’
• Emotional empathy - ‘‘an emotional response to . . . emotional responses of others’’’ (Lawrence,
Shaw, Baker, Baron-Cohen, & David, 2004, p. 911).
• In SLA, learners with higher Cognitive Empathy has been shown to have better attainment, and vice
versa.
• Instrument: Baron-Cohen and Wheelwright’s (2004) Empathy Quotient questionnaire.
FINDINGS
• A total of 2,158 multilinguals (1589 females, 457 males) completed a language use
questionnaire and the Baron-Cohen/Wheelwright EQ questionnaire, focusing on Cognitive
Empathy.
• Participants knowing more languages did not score higher on cognitive empathy than those
knowing fewer – knowing more languages alone does not enhance Cognitive Empathy.
• Participants who use multiple languages more frequently scored significantly higher on
cognitive empathy.
• Participants who habitually codeswitch between multiple language showed a stronger effect
on cognitive empathy than mere proficiency in multiple languages.

• Separately Dewaele and others investigated CS and emotions.

TOLERANCE OF AMBIGUITY
• TA is ‘‘tendency to perceive ambiguous situations as desirable’’ (Budner 1962: 29).
• “TA refers to the way an individual (or group) perceives and processes information about
ambiguous situations when they are confronted by an array of unfamiliar, complex or
incongruent cues” (Furnham 1994: 403)
• => correlated with Openness (behaviour: wide interests, imaginative & insightful, linked to
activity in dorsolateral cortex; considered primarily a cognitive trait) & Rigidity (inflexibility,
difficulty making transitions, adherence to set patterns, linked to deficit of the executive
functions (frontal lobe).

• In SLA, some studies have shown that good language learners are more tolerant of
ambiguity, though it remains a controversial issue.
• Multilingual Use questionnaire with 18 questions related to sociobiographical
background, frequency of codeswitching and attitudes towards CS etc.

• Adapted version of Herman’s (2010) Tolerance of Ambiguity questionnaire

• N = 2158 (1589 females, 457 males)
EFFECT OF MULTILINGUALISM ON TOLERANCE OF
AMBIGUITY
F =2.33, P < 0.041 ETA2 : 0.006
2,7

2,65
Tolerance of ambiguity

2,6

2,55

2,5

2,45
Mono Bi Tri Quadri Penta Sexta+
Number of languages known
Effect of TA on self-reported frequency of
Code-switching (ANOVAs)

2,5

2
Frequency of CS

1,5 Friends
Family
Strangers
1 Colleagues

0,5

0
Low Medium High
Tolerance of ambiguity
EFFECT OF TA ON ATTITUDES TOWARD
CODE-SWITCHING
1,6

1,4

1,2
Negative attitude towards CS

1
Incompetence
Annoying
0,8
Arrogant
BotheredDifference
0,6

0,4

0,2

0
Low Medium High
Tolerance of ambiguity
RESULTS
Participants who know more languages score high on TA.
TA not linked to proficiency
TA not linked frequency of CS (!)
TA linked to attitudes towards CS - High TA less likely to view CS
negatively or to be bothered by being different!
CS AND SELECTIVE ATTENTION, AND CREATIVITY
• with Anatoliy Kharkhurin

• Kharkhurin - effect of speaking several languages on an individual’s creative capacities.

• Individuals who know many different languages have better/enhanced selective attention, i.e. control
and separation
• Selective attention is crucial to creativity, i.e. divergent thinking
• Using the Stroop task, Kharkhurin revealed that bilinguals who are better at focusing on relevant
information i.e. selective attention, tend to also activate a larger number of possible solutions to a
problem (i.e., generative capacity).
• It also revealed that bilinguals with high language skills may utilize the inhibition mechanism of
selective attention to enhance the extraction of innovative and useful ideas (i.e., innovative capacity)
presumably by suppressing the interference of the ideas that fail to satisfy task requirements.
• Kharkhurin, 2011, made a logical though speculative conclusion that habitual CS where multiple
languages are simultaneously activated may hinder selective attention and therefore may have
negative impact on creative performance.
CS AND SELECTIVE ATTENTION, AND CREATIVITY
• The performance of 166 multilingual college students in UAE (59 male and 107 female, all Arabic-
English bilingual with various other languages) with different code-switching behaviors and attitudes
was tested on a battery of creativity and cognitive measures.
• Participants’ creative abilities were assessed using the Abbreviated Torrance Test for Adults (ATTA,
Goff & Torrance, 2002). It has three paper and pencil activities.
• In Activity 1, participants were asked to suppose that they could walk on air or fly, and then to
identify the troubles that they might encounter. This activity provided verbal fluency and originality
scores.
• In Activity 2, participants were presented with two abstract and incomplete figures and were asked
to draw pictures with these figures and to attempt to make these pictures as unusual as possible.
This activity provided figural fluency, originality, and elaboration scores.
• In Activity 3, the participants were presented with a group of nine triangles arranged in a 3 x 3 matrix
and were asked to draw as many pictures or objects as they could using those triangles. This
activity provided figural fluency, originality, elaboration, and flexibility scores.
• .
• ATTA offers four scores of fluency, originality, elaboration, and flexibility.
• Fluency measures the ability to produce quantities of ideas, which are relevant to the task
instructions. The sum of fluency scores in all three activities provided a fluency raw score.
• Originality measures the ability to produce uncommon ideas, or ideas that are totally new or
unique. The sum of originality scores in all three activities provided an originality raw score.
• Elaboration measures the ability to embellish ideas with details. The sum of elaboration
scores in Activities 2 and 3 provided an elaboration raw score.
• Flexibility measures the ability to process information or objects in different ways, given the
same stimulus. A flexibility raw score was obtained from Activity 3.
• The raw scores for fluency, originality, elaboration, and flexibility obtained in the test were
subsequently transformed into scaled norm-referenced scores by the recommended
procedure (Goff & Torrance, 2002) which took age-related norms into account.
• Participants’ fluid intelligence (Gf) was assessed by a standard Culture Fair Intelligence Test battery
(CFIT, Cattell, 1973), which uses nonverbal stimuli to assess intelligence in such a way that the
influence of verbal fluency, culture, and educational level has the least effect possible.

• Selective attention was assessed by a version of the standard Eriksen flanker task. Participants
were first presented with a fixation cross for 500 ms, which was immediately followed by a horizontal
array of five equally sized and spaced arrows for 1700 ms. The array was 14.87 cm wide and 1.16
cm high. The stimuli were presented in black on white background using 19” flat monitor.
Participants were instructed to attend to the central arrow and ignore the four flankers. They were to
press the left key for a left facing central arrow and the right key for a right facing central arrow. The
flanking arrows either all pointed in the same direction as the target arrow, or they all pointed in the
opposite direction. The trials on which the flanking arrows pointed in the same direction as the target
arrow were the congruent trials; the trials in which they pointed in the opposite direction were the
incongruent trials. Subjects received a total of 80 trials (40 congruent and 40 incongruent ones) in a
random order, requiring an equal number of left or right responses.
 ATTA ACTIVITY 2

KHARKHURIN & LI WEI: CS & CREATIVITY 27

 ATTA ACTIVITY 2

KHARKHURIN & LI WEI: CS & CREATIVITY 28

 ATTA ACTIVITY 3

KHARKHURIN & LI WEI: CS & CREATIVITY 29

 ATTA ACTIVITY 3

KHARKHURIN & LI WEI: CS & CREATIVITY 30

RESULTS
• The study revealed both effects: code-switching was found to
weaken an individual’s selective attention, yet at the same time, it
facilitated certain creative capacities and had no overall negative
consequences for creativity.
• Specifically, CS for special communication purposes was found to
be detrimental for selective attention, but not for creativity.
• On the other hand, CS induced by a particular emotional state and
by a lack of specific vocabulary in a target language appeared to
relate to increase in both generative and innovative capacities.
• Participants who code switch to achieve special communicative effect revealed
lower selective attention capacity. In these cases, they are likely to consider
several alternatives in different languages to select a lexical entry that
communicates their message in the best possible manner. The success of this
process partially relies on the ability to keep the entries in several languages
active.
• Code switchers seem to be unlikely to focus on one language and suppress the
other; that is, they should be less readily involved in interference suppression.
This explains the findings that individuals who code switch to achieve special
communication effect might be less involved in habitual interference
suppression and therefore showed poorer selective attention performance.
• At the same time, this performance was not related to any impairment in creative functioning.
Despite the fact that CS could be detrimental to selective attention, those individuals who
code switch to say something unusual do not suffer from limited selective control. They may
code switch to exercise their verbal creative capacity, which compensates for the lack of
selective attention.
• In an attempt to convey the message with special communication effects, they deliberately
code switch to achieve an expressive and creative performance. This idea is supported by our
finding of a clear tendency for bilinguals to code switch in order to convey a message better
and with more precision. This finding is consistent with the arguments put forward by
sociolinguists who suggest that one of the primary motivations for CS by bilingual speakers is
to convey messages more effectively, often through reiteration and elaboration in different
languages (e.g. Gumperz, 1982). Moreover, there is an argument that it is the contrast in
language choice that is built by the act of CS rather than the directionality of language choice
(e.g. Li, 2005).
DISCUSSION OF THE FINDINGS OF THE
SELECTIVE ATTENTION AND CREATIVITY TESTS
• When we talk about the relationship between CS and selective attention, we
assume that bilinguals employ this capacity only to suppress one language
while speaking the other. This may not be the case in the CS mode, as one still
has to select elements from different languages and mix them in order to
produce grammatical and meaningful utterances. It could be argued that
habitual code-switchers exercise more selective control when they are engaged
in CS, albeit at a much fine-grained level.
• This consideration could explain the findings of no selective attention
performance differences between participants who indicated that they code
switch in a particular emotional state, to convey a message better, and due to
the lack of a word in a target language and those who do not code switch for
these reasons.
• Definitely, the impact of CS on selective attention appears to have no negative
consequences for one’s creative capacity. Different CS conditions were found to
be facilitatory for creative functioning, including both generative and innovative
capacities.
• In particular, emotion-triggered and culturally-specific concept/word-triggered
CS results in higher scores in creative capacity measure.
FUTURE RESEARCH
• CS, the defining characteristic behavior of bilingual speakers, has not been
systematically studied in cognitive psycholinguistic research.
• We have tried to investigate the consequence of codeswitching on social cognition
of multilinguals, and the relationship between CS, selective attention and creativity,
partly to counter the negative views of CS by educators and others.
• Further research is required to examine other aspects of social cognition and how
CS contributes to creativity; does CS require more or less selective attention; what
cognitive mechanisms and psychological states may have an impact on the
relationships between CS, selective attention and creativity, and different types of
codeswitching (motivations, modalities).
• A model of cognitive control of CS is needed, that does not focus exclusively on
differentiation/separation/deactivation.
• Moreover, individual variations need to be taken seriously and investigated
systematically in psycholinguistics studies of CS, going beyond comparisons
between monolinguals and bilingualism, recognizing the diversity and the ‘ecology’
of different types of bilingual and multilingual language users.
• 2014 A control process model of codeswitching (with David Green). Language,
Cognition and Neuroscience 29,4: 499-511
DOI:10.1080/23273798.2014.882515
• 2014 The role of codeswitching in bilingual creativity and selective attention
(with Anatoliy Kharkhurin). International Journal of Bilingual Education and
Bilingualism. DOI: 10.1080/13670050.2014.884211
• 2013 Is multilingualism linked to a higher tolerance of ambiguity? Bilingualism:
Language and Cognition. 16.1: 231-240 (with Jean-Marc Dewaele)
• 2012 Multilingualism, empathy and multicompetence. International Journal of
Multilingualism 9.4: 352-366 (with Jean-Marc Dewaele)
THANK YOU
• Li Wei
• [email protected]
• Birkbeck College, University of London