Content uploaded by Paula Rubio-Fernandez
Author content
All content in this area was uploaded by Paula Rubio-Fernandez on Sep 24, 2014
Content may be subject to copyright.
RESEARCH REPORT
Reasoning About Other People’s Beliefs: Bilinguals Have an Advantage
Paula Rubio-Ferna´ndez and Sam Glucksberg
Princeton University
Bilingualism can have widespread cognitive effects. In this article we investigate whether bilingualism
might have an effect on adults’ abilities to reason about other people’s beliefs. In particular, we tested
whether bilingual adults might have an advantage over monolingual adults in false-belief reasoning
analogous to the advantage that has been observed with bilingual children. Using a traditional false-belief
task coupled with an eye-tracking technique, we found that adults in general suffer interference from their
own perspective when reasoning about other people’s beliefs. However, bilinguals are reliably less
susceptible to this egocentric bias than are monolinguals. Moreover, performance on the false-belief task
significantly correlated with performance on an executive control task. We argue that bilinguals’ early
sociolinguistic sensitivity and enhanced executive control may account for their advantage in false-belief
reasoning.
Keywords: bilingual cognition, theory of mind, false-belief reasoning, executive control, eye tracking
Experience can have powerful effects on cognitive abilities. For
example, London taxi drivers have enlarged regions of the hip-
pocampus involved in spatial navigation, and the extent of this
enlargement correlates positively with the time spent taxi driving
(Maguire et al., 2003). Action video-game playing enhances visual
selective attention (Green & Bavelier, 2003). Canadian postal
workers who regularly process letters and digits together in postal
codes have better integration of letter and digit processing than do
people without this experience (Polk & Farah, 1998). Experience
with a variety of cognitive tasks across the life span increases
cerebral reserve, reflecting the protective effect of intellectual
exercise against cognitive decline with aging (Valenzuela & Sa-
chdev, 2006).
Bilingualism provides experiences that can lead to various cog-
nitive effects but with mixed outcomes (Bialystok, 2009; Bialystok
& Craik, 2010). On the negative side, bilingual children have a
smaller vocabulary in each language than do their monolingual
peers (Oller & Eilers, 2002), and bilingual adults underperform in
word retrieval tasks relative to monolinguals, partly because of the
interference of their other language (Kaushanskaya & Marian,
2007). On the positive side, bilingual language production requires
constant monitoring of the target language in order to minimize
interference from the competing language. This requires the exer-
cise of executive control, which in turn should strengthen a per-
son’s executive control system (Bialystok, 2009; Bialystok &
Craik, 2010; cf. Costa & Santesteban, 2004, for an alternative
model of bilingual speech production without inhibitory control).
Bilinguals’ advantage in executive control tasks has been found in
children as well as in young and older adults (Bialystok, Craik,
Klein, & Viswanathan, 2004; Carlson & Meltzoff, 2008; Costa,
Herna´ndez, & Sebastia´n-Galle´s, 2008).
Bilingual children also perform better in false-belief tasks than
do their monolingual peers (Goetz, 2003; Kova´cs, 2009). False-
belief tasks have been a key test of social cognition develop-
ment—as well as being used in the diagnosis of autism and related
disorders—for more than 2 decades (Onishi & Baillargeon, 2005;
Wellman, Cross, & Watson, 2001; Wimmer & Perner, 1983). In
the classic Sally–Anne task (Baron-Cohen, Leslie, & Frith, 1985),
children are presented with two puppets, Sally and Anne, who are
playing with a toy. When they finish playing, they put the toy into
a box and Anne leaves the scene. While she is away, Sally moves
the toy to a different box. When Anne comes back, the child is
asked where she will look for the toy. Monolingual children
around the age of 4 years typically answer correctly that she will
look for the toy in the original container. However, younger
monolingual children as well as some autistic individuals tend to
respond according to their own knowledge of the situation, thus
failing to show an appreciation of Anne’s false belief about the
location of the toy. In contrast, bilingual children as young as 3
years have shown a precocious success in false-belief tasks (Goetz,
2003; Kova´cs, 2009).
This article was published Online First August 29, 2011.
Paula Rubio-Ferna´ndez and Sam Glucksberg, Department of Psychol-
ogy, Princeton University.
This research was supported by a British Academy Postdoctoral Fellow-
ship awarded to the first author and by a Marie Curie Outgoing Interna-
tional Fellowship awarded to both authors (Project 022149). We thank
Ellen Bialystok and Marianne Gullberg for their advice on the distinction
between bilinguals and monolinguals and to Natalie Sebanz for her sug-
gestions for the Simon task. Special thanks to Pedro Montoro and Mark
Pinsk for their technical help. Thanks to Albert Costa, Robyn Carston, Kay
Deaux, Susanne Grassmann, and Boaz Keysar for their comments on early
drafts of the article.
Correspondence concerning this article should be addressed to Paula
Rubio-Ferna´ndez, Department of Psychology, Princeton University, Green
Hall, Princeton, NJ 08540. E-mail: prubio@princeton.edu
Journal of Experimental Psychology: © 2011 American Psychological Association
Learning, Memory, and Cognition
2012, Vol. 38, No. 1, 211–217 0278-7393/11/$12.00 DOI: 10.1037/a0025162
211
The extent to which young children may fail traditional false-
belief tasks because they are not able to appreciate another per-
son’s beliefs or because they are not able to inhibit their own
knowledge of the situation has been a recurrent theme in the
developmental literature (Perner & Lang, 1999; Wellman et al.,
2001). Bilingual children’s advantage in traditional false-belief
tasks might well be related to both factors. Bilingual children must
develop an early sociolinguistic sensitivity to the language knowl-
edge of their interlocutors because they must use their languages
accordingly (Genesee, Boivin, & Nicoladis, 1996; Genesee, Nico-
ladis, & Paradis, 1995). Even though this fundamental aspect of
bilingual children’s experience has not been investigated in the
context of false-belief reasoning, their awareness that other people
do not always speak the same languages as they do might be an
early form of appreciating that other people might have a different
perspective from their own. Moreover, this early form of perspec-
tive taking is combined with an early developed executive control
system that is necessary to focus on the target language and avoid
interference from the contextually inappropriate linguistic system
(Kova´ cs & Mehler, 2009). Their advanced executive control
would help bilingual children inhibit their own perspective in
false-belief tasks (Kova´cs, 2009). In general, language learning
itself can promote the development of children’s theory of mind—
their understanding of other people’s mental states and intentions
(Pyers & Senghas, 2009).
Although it has been established that at around 4 years of age
children are able to pass standard false-belief tasks (Wellman et
al., 2001), more recent studies using eye-gaze measures to monitor
infants’ expectations have suggested that even 15-month-old in-
fants might be able to succeed in nonverbal versions of the false-
belief task (Onishi & Baillargeon, 2005; Southgate, Senju, &
Csibra, 2007). Nevertheless, although infants might be able to pass
nonverbal versions of the false-belief task and older children
perform optimally in standard false-belief tasks, adults’ perfor-
mance on a variety of judgment tasks can still be affected by the
curse of knowledge—a tendency to be biased by their own knowl-
edge of the situation (Birch & Bloom, 2007; Camerer, Lowenstein,
& Weber, 1989; Kelley & Jacoby, 1996; Keysar, 1994). It is
therefore possible that, using sufficiently fine grained measures,
adults might reveal an egocentric bias in their performance on a
standard false-belief task.
We investigated the false-belief reasoning abilities of adult
bilinguals relative to monolinguals using a standard false-belief
task coupled with an eye-tracking technique. Monitoring partici-
pants’ eye movements during performance in the traditional Sally–
Anne task should enable us to observe the effects of the curse of
knowledge in adult false-belief reasoning. If adult participants first
consider the container where they know the toy is before taking the
perspective of the mistaken story character, then this would im-
plicate an egocentric bias. Recent eye-tracking research has shown
that adult participants do suffer from an egocentric bias when
performing referential communication tasks (Epley, Morewedge,
& Keysar, 2004; Keysar, Barr, Balin, & Brauner, 2000; Keysar,
Lin, & Barr, 2003). For example, in following the instruction
“Move the small candle,” participants tend to initially consider the
smallest of three candles on display, even though that particular
candle is not visible to the speaker and therefore cannot be the
intended target.
If eye movements reveal an egocentric bias in adult false-belief
reasoning, do bilingual participants suffer less from the curse of
knowledge than do monolinguals? Given their sociolinguistic sen-
sitivity to the language background of their interlocutors as well as
their enhanced executive control, bilingual adults may maintain an
advantage in false-belief reasoning analogous to the advantage
found in bilingual children (Goetz, 2003; Kova´cs, 2009).
Method
Participants
We tested 46 undergraduates at Princeton University: 23 bilin-
guals and 23 monolinguals. All participants were to some extent
familiar with a second language. For the purpose of the study, we
were interested in distinguishing those participants who had been
regularly using two languages for most of their lives from those
who had not. The two main criteria for classifying our participants
as bilinguals or monolinguals were (a) for how many years they
had known a second language and (b) how regularly they had had
to switch between their first and second languages in that time. To
obtain this information, we used the Language History Question-
naire (Gullberg & Indefrey, 2003).
We established that the bilinguals in our sample should have
learned a second language (L2) before age 9 and have been using
it regularly for 10 years or more. The mean age of acquisition of
the L2 was 3, and the mean number of years of use of the L2 was
16. The bilinguals in our sample had learned their L2 in (a) a
bilingual household (12 participants), (b) a monolingual household
using a language foreign to the country of residence (six partici-
pants), or (c) a bilingual school (five participants). The bilinguals
in our group would have therefore had to switch languages daily or
almost daily for a number of years. The L2s in our sample were
Spanish (5), Chinese (8), Hindi (1), Korean (2), Russian (1),
Hebrew (2), French (2), Farsi (1), and Japanese (1).
1
Because all participants in the study were Princeton students,
both bilingual and monolingual participants were recruited from a
selective pool of high-performance students who had been ac-
cepted into university under rigorous academic criteria. Given our
samples, no important differences in IQ or verbal ability would be
expected. The mean age was 19.7 years for the bilinguals and 19.4
1
Having a collectivistic cultural background might give people an
advantage in perspective taking (Wu & Keysar, 2007). However, this
advantage seems to be observable only in individuals who were born and
brought up in collectivistic countries. Wu and Keysar (2007) failed to find
this effect in American Chinese students, whereas an advantage was found
when comparing Chinese students who had arrived in the United States
within the last year and American students (B. Keysar, personal commu-
nication, April 24, 2009). In our sample, 12 bilingual participants had an
Asian collectivistic background, although five of those participants had
been born in the United States. Three participants in our monolingual group
also had an Asian collectivistic background. It was not originally our
intention to investigate the effects of cultural background on false-belief
reasoning. However, given the considerable number of Asian participants
in our sample, we decided to further analyze our data comparing Asian
collectivistic participants (n⫽15) and non-Asian individualistic partici-
pants (n⫽31) alongside the main comparisons between monolinguals and
bilinguals.
212 RUBIO-FERNA
´NDEZ AND GLUCKSBERG
years for the monolinguals. Of the 23 bilingual participants, 15
were female; of the 23 monolinguals, 12 were female. All partic-
ipants had either been born in the United States (65% of bilinguals
and 87% of monolinguals) or moved to the United States or to
another English-speaking country by age 12. Crucially, at the time
they took part in the study, English was the dominant language of
all participants, whether bilingual or monolingual.
Sally–Anne Task
Procedure. The false-belief task was an extended computer
version of the classic Sally–Anne task. Two kindergarten charac-
ters, Sally and Anne, interacted in an animated cartoon. Each child
had a favorite toy and a container in which she kept it before going
home every day. Participants were familiarized with the setting of
the story in two warm-up trials before they were tested in the
following two different conditions:
False-belief (FB) condition: Anne puts her doll in her basket
and goes home. While she is away, Sally moves Anne’s doll
from the basket to the box. When Anne comes back the next
day, participants are asked “Where will she look for her
doll?”
True-belief (TB) baseline: Sally puts her horse in her box and
goes home. Anne was also going to put her doll in her basket,
but because her basket is getting full, she decides to put it in
the box. When Anne comes back the next day, participants
are asked “Where will she remember to go and find her
doll?”
2
The FB condition is the standard Sally–Anne scenario in which
Anne erroneously believes that her doll is in the basket where she
left it. The TB condition served as a baseline for FB because
Anne—like the participants—is correct about the location of her
doll. Because the TB baseline was always presented after the FB
condition, we used a different comprehension question in the TB
condition in order to avoid facilitation due to repetition of the
question. Given the greater syntactic complexity of the TB ques-
tion, however, longer response times (RTs) were expected in the
baseline condition. Because predictability was a potential issue for
finding evidence of an egocentric bias with a simple task, partic-
ipants were tested in only one trial per condition, with the first
experimental trial always corresponding with the FB condition. In
order to separate the trials clearly, a short clip with an adaptation
of a traditional children’s song was played after each scene.
Unlike the young children for whom the task was originally
designed, adult participants were expected to respond to the FB
question correctly. Any evidence of the curse of knowledge in this
population should be observed only in their eye movements and
RTs.
Participants were given standard instructions that described their
role as a control group in an experiment aimed at children. Par-
ticipants watched the Sally–Anne cartoon on a computer screen
and listened to the accompanying story, which lasted approxi-
mately 4 min. At different points in the story they were asked a
comprehension question, which they answered using one of two
labeled keys on the computer keyboard. The position of the box
and basket keys was symmetric to that of the containers on the
screen. In order to avoid any possible bias related to a differential
left- and right-hand dominance, participants were asked to press
the response keys with their dominant hand.
In order to maximize the extent of eye movements, immediately
before each comprehension question the containers disappeared
from the scene while the narrative continued. Then Anne reap-
peared on the lower center of the screen to make sure that all
participants were fixating on the same point when the two con-
tainers reappeared in the top corners of the screen (see Figure 1).
The presentation of the display where the containers reappear on
the screen was taken as the onset for eye fixations. The containers
reappeared at the onset of the verb in the questions in order to
obtain an accurate measure of the initial stages of perspective
taking. A fixation was defined as an eye movement that remained
on one of the target areas of interest for a minimum of 100 ms. The
onset for RTs corresponded with the offset of the question.
Results.
Gaze direction. The first measure that we used was direction
of first fixation in the FB condition. That is, on which target
participants first fixated when the containers reappeared on the
screen during the FB question. Overall, 19 participants (41.3%)
looked first at the correct container where the child had left her toy,
whereas 26 participants (56.5%) revealed an egocentric bias, look-
ing first at the container where they knew the toy was before
correcting this tendency and focusing their gaze on the correct
container. One monolingual participant did not fixate on either
container before responding to the question.
Considering bilinguals and monolinguals separately, 13 bilin-
gual participants (56.5%) gazed directly at the correct container,
whereas only six monolinguals (26.1%) did so, with the majority
of monolingual participants revealing an egocentric bias (see Fig-
ure 2). This difference is reliable,
2
(1, N⫽45) ⫽3.94, p⬍.048.
Whereas in the FB condition the wrong container represented
the participants’ privileged knowledge of the location of the toy
and the correct container represented the character’s perspective,
in the TB baseline there was no competition between the two
targets. It is interesting that in the TB condition, 21 bilingual
participants (91.3%) and 18 monolingual participants (78.3%) first
fixated on the correct target, which did not result in a reliable
difference,
2
(1, N⫽46) ⫽1.52, p⫽.218.
Fixation latency. The second eye-tracking measure was fix-
ation latencies on the correct container in the FB condition. That is,
how long it took participants to first fixate on the correct target
when the containers reappeared on the screen at the end of the FB
question. Data points that were above 2.5 standard deviations over
the group’s mean fixation latency on the correct target were
discarded as outliers. Outlying responses (two in the bilingual data
set and two in the monolingual data set) as well as data points that
did not reveal a fixation on the correct target (one in the bilingual
data set and three in the monolingual data set) were replaced by the
group’s mean fixation latency on the correct target without the
outlying responses.
Given the larger number of first fixations on the wrong con-
tainer among monolingual participants, bilingual participants were
on average faster at first fixating on the correct target (mean 619
ms) than were the monolingual participants (mean 812 ms; see
2
Given the syntactic ambiguity of the TB question, the intended reading
was stressed in the recording of the materials.
213
BILINGUAL ADVANTAGE IN FALSE-BELIEF REASONING
Figure 3). This difference was reliable, t(44) ⫽2.07, p⬍.045,
thus confirming the better performance of the bilingual group.
Response time. The third measure was RTs for the FB and TB
questions. Data points that were above 2.5 standard deviations
above the group’s mean RTs for each condition were discarded as
outliers. Outlying responses (one bilingual) as well as data points
that revealed a second press on the response key (i.e., the partic-
ipant had first pressed the response key before the offset of the
question; one bilingual and one monolingual) were replaced by the
group’s mean RT in that condition without the outlying responses.
Given that we used a different question for the FB condition and
the TB baseline—with the second question eliciting much longer
RTs than the first one because of its greater syntactic complexity
(see Figure 4)—we computed the relative difference between
participants’ RTs to the TB question and the FB question over the
TB baseline. This relative measure of RT did not reveal a reliable
difference between the two groups, t(44) ⫽1.52, p⫽.135,
suggesting that RTs might be less sensitive than eye-movement
measures.
Regarding baseline performance, RTs to the TB question did not
differ reliably for bilinguals (M⫽1,692 ms) and monolinguals
(M⫽1,563 ms), t(44) ⫽1.38, p⫽.175. We take the bilinguals’
comparable RTs in the TB baseline as evidence that this group was
not overall superior to the monolingual group in the context of this
experiment.
3
RTs to the FB question did not differ reliably for bilinguals
(M⫽916 ms) and monolinguals (M⫽1,042 ms), t(44) ⫽0.891,
p⫽.378. RTs therefore seem to be less sensitive than eye-tracking
measures, which did reveal a reliable difference between bilinguals
and monolinguals in the FB condition. When we collapsed the RTs
for bilinguals and monolinguals and divided them between those
who first fixated on the correct container (M⫽812 ms) and those
who did not (M⫽1,097 ms), we also observed a reliable differ-
ence, t(44) ⫽2.05, p⬍.048. Given that this pattern of RTs was
driven by accuracy of first fixation and significantly more bilin-
guals than monolinguals first fixated on the correct container, it is
likely that with a larger sample of participants, a reliable difference
in favor of the bilingual group might also be observed in the RT
measure.
4
Simon Task
Procedure. After completing the Sally–Anne task, partici-
pants were given a version of the Simon task in order to assess
their level of executive control (O’Leary & Barber, 1993). In this
task, participants had to press a right-hand key when they saw the
word RIGHT on the computer screen and a left-hand key when
they saw the word LEFT. In the baseline condition, the location
of the words on the computer screen matched the actual words
3
It has been documented that English–Spanish bilinguals, for example,
might show parsing preferences when processing certain syntactically
ambiguous constructions because of interference from their other language
(see Desmet & Duyck, 2007, for a review). This could suggest that at least
some of the bilinguals in our sample might have had problems parsing the
TB question—even though the intended reading was stressed in the re-
cording. However, recent studies have shown that the erosion of parsing
strategies in the first language occurs only after prolonged immersion in an
L2 environment (Dussias, 2004; Dussias & Sagarra, 2007). Because all
bilingual participants in our study had English as their dominant language
and had been living in an English-speaking country for the last 10 years or
more, we assumed that their parsing preferences would have been compa-
rable to those of the monolinguals in our study.
4
The comparison between Asian collectivistic and non-Asian individu-
alistic participants did not reveal any reliable results in the analyses of gaze
direction, fixation delay, and RT in the Sally–Anne task.
Figure 1. Display from the Sally–Anne cartoon corresponding with the
question “Where will she look for her doll?”
40
50
60
70
80
90
100
*
0
10
20
30
40
50
BILINGUALS MONOLINGUALS
Percentage of Participants
Figure 2. Percentage of participants (⫾standard error of the means) who
first fixated on the correct container in the false-belief condition.
ⴱ
p⬍
.049, two-tailed.
500
600
700
800
900
1000 *
0
100
200
300
400
BILINGUALS MONOLINGUALS
Fixation latency (ms)
0
Figure 3. Mean fixation latencies (⫾standard error of the means) on the
correct container in the false-belief condition.
⫹
p⬍. 045, two-tailed.
214 RUBIO-FERNA
´NDEZ AND GLUCKSBERG
(e.g., the word RIGHT appeared on the right-hand side). In con-
trast, in the experimental condition the location of the words was
switched (e.g., the word RIGHT appeared on the left-hand side). In
this second condition, participants had to overcome the impulse to
respond according to the location of the word rather than to the
word’s meaning.
5
The task consisted of 50 trials: 10 practice trials and 40 critical
trials. All practice trials were in the baseline condition and were
not analyzed. Of the 40 critical trials, 20 corresponded with the
baseline condition and 20 with the experimental condition. The
words appeared in black capital letters (in 54-point Arial font)
against a white background on a 15-in. screen. In between trials,
the screen remained blank for 600 ms, and then a blue cross
appeared in the center of the screen as a fixation point for 1,000
ms. A pseudorandom sequence was generated to present the trials
so that the same response would not occur in more than two
consecutive trials. One monolingual participant was removed from
the data set because of her high error rate (25% of responses; 45%
in the critical condition).
Results. As in previous studies, bilingual participants showed
less interference in the experimental condition relative to the
baseline condition than did monolinguals (see Figure 5). A 2 ⫻2
analysis of variance revealed a significant main effect of condition,
F(1, 43) ⫽48.68, p⬍.001. The Condition ⫻Group interaction
was also significant, F(1, 43) ⫽4.70, p⬍.037, reflecting the
superior performance of the bilingual participants in the executive
control task. Performance on the baseline condition was compa-
rable for the two groups.
6
Correlation Between the Sally–Anne Task
and the Simon Task
For the correlation between the two tasks, we computed fixation
duration on the wrong container in the FB condition of the Sally–
Anne task. That is, for how long participants fixated on the wrong
container from the point when the containers reappeared on the
screen at the end of the FB question until they responded to the
question. This measure therefore includes first and later fixations.
One bilingual participant was eliminated from the analysis because
of calibration problems that affected this measure, and one mono-
lingual participant who did not make any eye movements during
the processing of the FB question was eliminated. Fixation dura-
tion on the wrong container did not reveal a reliable difference
between bilinguals (M⫽206 ms) and monolinguals (M⫽262
ms), t(42) ⫽0.961, p⫽.342.
We took fixation duration on the wrong container of the FB
condition as a measure of egocentric bias, and as such we expected
it to positively correlate with participants’ level of executive
control. That is, the larger the fixation duration on the wrong
container in the FB condition of the Sally–Anne task, the larger the
difference between the experimental and the baseline conditions in
the Simon task. We calculated the participants’ level of executive
control as the difference between their RTs in the baseline and the
experimental conditions (M
bilinguals
⫽26 ms; M
monolinguals
⫽49
ms). The correlation between the Sally–Anne task and the Simon
task was indeed significant for bilinguals, r(20) ⫽.513, one-tailed
p⬍.008, and for monolinguals, r(19) ⫽.529, one-tailed p⬍.008.
Of the 43 participants who were included in the correlation anal-
ysis, 12 (seven bilinguals and five monolinguals) did not fixate at
all on the wrong container while processing the FB question.
Without these 12 participants (for whom fixation duration was 0
ms), the correlation between the Sally–Anne task and the Simon
task was still significant for bilinguals, r(13) ⫽.761, one-tailed
p⬍.001, and for monolinguals, r(14) ⫽.491, one-tailed p⬍.028.
Discussion
Our overall results show that adults suffer from an egocentric
bias in FB reasoning. Unlike in the case of young children,
however, this bias does not affect task performance: Adults always
answer the FB question correctly. However, with the use of an
eye-tracking technique, we observed that the majority of adults
momentarily consider the egocentric response before correcting
this tendency and taking the perspective of the story character.
These results are in line with previous studies revealing an ego-
centric bias in several different areas of cognition (Birch & Bloom,
2007; Camerer et al., 1989; Keysar, 1994; Keysar et al., 2003).
5
The version of the Simon task that we used also has elements of the
Stroop task, given the conflict between word meaning and word position.
We are grateful to Helen Bialystok for pointing out to us that both types of
task have shown bilingual advantages.
6
The comparison between Asian collectivistic and non-Asian individu-
alistic participants in the Simon task did not reveal a reliable difference
between the two groups.
1000
1200
1400
1600
1800
2000
Bilingual
0
200
400
600
800
FB TB
Response time (ms)
Monolingual
0
Figure 4. Mean response times (⫾standard error of the means) to the
false-belief (FB) question and the true-belief (TB) question.
350
400
450
500
550
BASELINE EXPERIMENTAL
Response time (ms)
Bilingual
Monolingual
**
Figure 5. Mean response times (⫾standard error of the means) in the
baseline and experimental conditions of the Simon Task.
ⴱ
p⬍.002,
two-tailed, for paired comparisons between conditions in the two groups.
215
BILINGUAL ADVANTAGE IN FALSE-BELIEF REASONING
Regarding differences between bilinguals and monolinguals, we
found that adult bilinguals suffer less from the curse of knowledge
than do monolinguals. This advantage in FB reasoning may be
maintained from an early age, because bilingual children outper-
form monolinguals in FB tasks (Goetz, 2003; Kova´cs, 2009). One
of the reasons for this advantage may be the bilinguals’ higher
level of executive control, which was confirmed in the present
study with the Simon task. Moreover, performance in the execu-
tive control task correlated with performance on the Sally–Anne
task. An enhanced executive control would help bilingual partic-
ipants inhibit their own knowledge of the situation in FB tasks,
making it easier to take the perspective of the story character
(Kova´cs, 2009).
A second factor that may account for the bilinguals’ advantage
in FB reasoning is their early sociolinguistic awareness of their
interlocutors’ language background (Comeau, Genesee, & Men-
delson, 2007; Genesee et al., 1996). Culture, for example, has been
found to have an effect on adult perspective-taking abilities in
referential communication tasks, with people from collectivistic
cultures suffering less interference from their own visual perspec-
tive than do people from individualistic cultures (Wu & Keysar,
2007). If the interdependence that is characteristic of collectivistic
cultures can make Chinese university students better at
perspective-taking tasks than American students, having to mon-
itor for the language background of their interlocutors from an
early age might also make bilinguals better perspective takers than
monolinguals.
Future research should determine whether the advantage in FB
reasoning that we have observed in adult bilinguals relative to
monolinguals is maintained throughout development from early
childhood (Goetz, 2003; Kova´cs, 2009). Bilinguals’ advantage in
executive control tasks has indeed been found in children, young
adults, and older adults (Bialystok et al., 2004; Carlson & Melt-
zoff, 2008; Costa et al., 2008). This advantage could in principle
help bilinguals maintain an advantage in FB reasoning during
development. Moreover, a recent study has shown that the
perspective-taking abilities used in referential communication
tasks continue to develop into late adolescence in monolinguals
(Dumontheil, Apperly, & Blakemore, 2010). It remains an empir-
ical question whether the perspective-taking advantage that bilin-
gual children have shown in language-switch situations, for exam-
ple (Comeau et al., 2007; Kova´cs, 2009), would also be maintained
throughout development and perhaps enhance their FB reasoning
abilities throughout the life span.
Overall, our results support previous research suggesting that
bilingualism is one of those life experiences that leads to general
cognitive effects (Bialystok, 2009; Bialystok & Craik, 2010).
Although these effects have been mixed in other areas of cogni-
tion, the effect of bilingualism on FB reasoning is a positive one.
References
Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child
have a “theory of mind”? Cognition, 21, 37–46. doi:10.1016/0010-
0277(85)90022-8
Bialystok, E. (2009). Bilingualism: The good, the bad and the indifferent.
Bilingualism: Language and Cognition, 12, 3–11. doi:10.1017/
S1366728908003477
Bialystok, E., & Craik, F. I. M. (2010). Cognitive and linguistic processing
in the bilingual mind. Current Directions in Psychological Science, 19,
19–23. doi:10.1177/0963721409358571
Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004).
Bilingualism, aging, and cognitive control: Evidence from the Simon
task. Psychology and Aging, 19, 290–303. doi:10.1037/0882-
7974.19.2.290
Birch, S. A. J., & Bloom, P. (2007). The curse of knowledge in reasoning
about false beliefs. Psychological Science, 18, 382–386. doi:10.1111/
j.1467-9280.2007.01909.x
Camerer, C., Lowenstein, G., & Weber, M. (1989). The curse of knowl-
edge in economic settings: An experimental analysis. Journal of Polit-
ical Economy, 97, 1232–1254. doi:10.1086/261651
Carlson, S. M., & Meltzoff, A. N. (2008). Bilingual experience and
executive functioning in young children. Developmental Science, 11,
282–298. doi:10.1111/j.1467-7687.2008.00675.x
Comeau, L., Genesee, F., & Mendelson, M. (2007). Bilingual children’s
repairs of breakdowns in communication. Journal of Child Language,
34, 159–174. doi:10.1017/S0305000906007690
Costa, A., Herna´ndez, M., & Sebastia´n-Galle´s, N. (2008). Bilingualism
aids conflict resolution: Evidence from the ANT task. Cognition, 106,
59–86. doi:10.1016/j.cognition.2006.12.013
Costa, A., & Santesteban, M. (2004). Lexical access in bilingual speech
production: Evidence from language switching in highly proficient bi-
linguals and L2 learners. Journal of Memory and Language, 50, 491–
511. doi:10.1016/j.jml.2004.02.002
Desmet, T., & Duyck, W. (2007). Bilingual language processing. Lan-
guage and Linguistics Compass, 1, 168–194. doi:10.1111/j.1749-818X
.2007.00008.x
Dumontheil, I., Apperly, I., & Blakemore, S. J. (2010). Online usage of
theory of mind continues to develop in late adolescence. Developmental
Science, 13, 331–338. doi:10.1111/j.1467-7687.2009.00888.x
Dussias, P. E. (2004). Parsing a first language like a second: The
erosion of L1 parsing strategies in Spanish-English bilinguals. Inter-
national Journal of Bilingualism, 8, 355–371. doi:10.1177/
13670069040080031001
Dussias, P. E., & Sagarra, N. (2007). The effect of exposure on syntactic
parsing in Spanish-English bilinguals. Bilingualism: Language and Cog-
nition, 10, 101–116. doi:10.1017/S1366728906002847
Epley, N., Morewedge, C. K., & Keysar, B. (2004). Perspective taking in
children and adults: Equivalent egocentrism but differential correction.
Journal of Experimental Social Psychology, 40, 760–768. doi:10.1016/
j.jesp.2004.02.002
Genesee, F., Boivin, I., & Nicoladis, E. (1996). Talking with strangers: A
study of bilingual children’s communicative competence. Applied Psy-
cholinguistics, 17, 427–442. doi:10.1017/S0142716400008183
Genesee, F., Nicoladis, E., & Paradis, J. (1995). Language differentiation
in early bilingual development. Journal of Child Language, 22, 611–
631. doi:10.1017/S0305000900009971
Goetz, P. (2003). The effects of bilingualism on theory of mind develop-
ment. Bilingualism: Language and Cognition, 6, 1–15. doi:10.1017/
S1366728903001007
Green, C. S., & Bavelier, D. (2003, May 29). Action video game modifies
visual selective attention. Nature, 423, 534–537. doi:10.1038/
nature01647
Gullberg, M., & Indefrey, P. (2003). Language Background Question-
naire: The dynamics of multilingual processing. Nijmegen, the Nether-
lands: Max Planck Institute for Psycholinguistics.
Kaushanskaya, M., & Marian, V. (2007). Bilingual language processing
and interference in bilinguals: Evidence from eye tracking and picture
naming. Language Learning, 57, 119–163. doi:10.1111/j.1467-
9922.2007.00401.x
Kelley, C. M., & Jacoby, L. L. (1996). Adult egocentrism: Subjective
experience versus analytic bases for judgment. Journal of Memory and
Language, 35, 157–175. doi:10.1006/jmla.1996.0009
216 RUBIO-FERNA
´NDEZ AND GLUCKSBERG
Keysar, B. (1994). The illusory transparency of intention: Linguistic per-
spective taking in text. Cognitive Psychology, 26, 165–208. doi:10.1006/
cogp.1994.1006
Keysar, B., Barr, D. J., Balin, J. A., & Brauner, J. S. (2000). Taking
perspective in conversation: The role of mutual knowledge in compre-
hension. Psychological Science, 11, 32–38. doi:10.1111/1467-
9280.00211
Keysar, B., Lin, S., & Barr, D. J. (2003). Limits on theory of mind use in
adults. Cognition, 89, 25–41. doi:10.1016/S0010-0277(03)00064-7
Kova´cs, A. M. (2009). Early bilingualism enhances mechanisms of false-
belief reasoning. Developmental Science, 12, 48–54. doi:10.1111/
j.1467-7687.2008.00742.x
Kova´cs, A. M., & Mehler, J. (2009). Cognitive gains in 7-month-old
bilingual infants. PNAS Proceedings of the National Academy of Sci-
ences of the United States of America, 106, 6556–6560. doi:10.1073/
pnas.0811323106
Maguire, E. A., Spiers, H. J., Good, C. D., Hartley, T., Frackowiak, R. S.,
& Burgess, N. (2003). Navigation expertise and the human hippocam-
pus: A structural brain imaging analysis. Hippocampus, 13, 250–259.
doi:10.1002/hipo.10087
O’Leary, M. J., & Barber, P. J. (1993). Interference effects in the Stroop
and Simon paradigms. Journal of Experimental Psychology: Human
Perception and Performance, 19, 830–844. doi:10.1037/0096-
1523.19.4.830
Oller, D. K., & Eilers, R. E. (2002). Language and literacy in bilingual
children. Clevedon, England: Multilingual Matters.
Onishi, K. H., & Baillargeon, R. (2005, April 8). Do 15-month-old infants
understand false beliefs? Science, 308, 255–258. doi:10.1126/
science.1107621
Perner, J., & Lang, B. (1999). Development of theory of mind and exec-
utive control. Trends in Cognitive Sciences, 3, 337–344. doi:10.1016/
S1364-6613(99)01362-5
Polk, T. A., & Farah, M. J. (1998). The neural development and organi-
zation of letter recognition: Evidence from functional neuroimaging,
computational modelling, and behavioural studies. PNAS Proceedings of
the National Academy of Sciences of the United States of America, 95,
847–852. doi:10.1073/pnas.95.3.847
Pyers, J. E., & Senghas, A. (2009). Language promotes false-belief under-
standing: Evidence from learners of a new sign language. Psychological
Science, 20, 805–812. doi:10.1111/j.1467-9280.2009.02377.x
Southgate, V., Senju, A., & Csibra, G. (2007). Action anticipation through
attribution of false belief by 2-year-olds. Psychological Science, 18,
587–592. doi:10.1111/j.1467-9280.2007.01944.x
Valenzuela, M. J., & Sachdev, P. (2006). Brain reserve and dementia: A
systematic review. Psychological Medicine, 36, 441–454. doi:10.1017/
S0033291705006264
Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-
of-mind development: The truth about false belief. Child Development,
72, 655–684. doi:10.1111/1467-8624.00304
Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and
constraining function of wrong beliefs in young children’s understand-
ing of deception. Cognition, 13, 103–128. doi:10.1016/0010-
0277(83)90004-5
Wu, S., & Keysar, B. (2007). Cultural effects on perspective taking.
Psychological Science, 18, 600–606. doi:10.1111/j.1467-
9280.2007.01946.x
Received July 16, 2010
Revision received July 1, 2011
Accepted July 5, 2011 䡲
217
BILINGUAL ADVANTAGE IN FALSE-BELIEF REASONING
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Experimental Psychology Learning Memory and Cognition
This content is subject to copyright. Terms and conditions apply.