ArticlePDF Available

Abstract and Figures

The universal cultural validity of the comparative ignorance hypothesis of ambiguity aversion was investigated. We also examined if cultural differences in choice preference were observed between Westerners, by using a sample of French participants, and Easterners by using a sample of Japanese participants. Furthermore, we investigated if differences in choice preference were observed within each cultural group. We conducted two experiments using the Ellsberg’s two-color problem. Results indicated that ambiguity aversion decreased in a non-comparative betting condition (Experiment 2) compared to a comparative betting conditions (Experiment 1), which supported the weak version of the comparative ignorance hypothesis. We also found that personal choice was more important for French than for Japanese participants. However French participants did not put as much weight on personal choice, as did Americans in previous studies.
Content may be subject to copyright.
Culture, Ambiguity Aversion and Choice in Probability
Judgments
Kuniko Adachi, Hiroshi Yama
Osaka City University, Japan
Jean-Baptiste Van der Henst, Hugo Mercier
Institut de Sciences Cognitives, France University of Pennsylvania, USA
Minoru Karasawa and Yayoi Kawasaki
Nagoya University, Japan Nihon University, Japan
The universal cultural validity of the comparative ignorance hypothesis of ambiguity
aversion was investigated. We also examined if cultural differences in choice
preference were observed between Westerners, by using a sample of French
participants, and Easterners by using a sample of Japanese participants. Furthermore,
we investigated if differences in choice preference were observed within each
cultural group. We conducted two experiments using the Ellsberg’s two-color
problem. Results indicated that ambiguity aversion decreased in a non-comparative
betting condition (Experiment 2) compared to a comparative betting conditions
(Experiment 1), which supported the weak version of the comparative ignorance
hypothesis. We also found that personal choice was more important for French
than for Japanese participants. However French participants did not put as much
weight on personal choice, as did Americans in previous studies.
Many recent cross-cultural studies have showed cultural differences in cognition
between Westerners and Easterners. Nisbett and colleagues (Nisbett, 2003; Nisbett,
Peng, Choi, & Norenzayan, 2001) have offered a model of two distinct modes of
cognitive processing by Westerners and East Asians (Chinese, Japanese, and Koreans);
these are, respectively, analytic cognition and holistic cognition. According to their
definition, analytic cognition involves detachment of the object from its context, a
tendency to focus on attributes of the object to assign it to categories, and a
preference for using rules about the categories to explain and predict the object’s
behavior. In contrast, holistic cognition involves an orientation to the context or field
as a whole, attention to relationships between a focal object and the field, and a
Correspondence concerning this article should be addressed to Kuniko Adachi, Urban-Culture Research
Center, Graduate School of Literature and Human Sciences, Osaka City University, Sugimoto, Sumiyoshi,
Osaka, 558-8585, Japan. E-mail: k-adachi@osa.att.ne.jp
This research has been supported by a CHORUS grant from the French Ministry of Research and the
Japanese Society for the Promotion of Sciences and by a grant-in-aid from the Japanese Society for the
Promotion of Sciences (No. 19202012; Project leader: Yukinori Takubo). We thank Craig Fox, Kimihiko
Yamagishi, David McCullough, Wai-Ling Lai and Yuriko Zemba for helpful comments on this article,
Yasuko Morinaga, Sayaka Kuranaka, Akira Mukai, Koji Kazai, Tetsuji Hirano, Nobutaka Doe, and Miwa
Nishioka, for their help in collecting the data.
THE INTERNATIONAL JOURNAL OF CREATIVITY & PROBLEM SOLVING
2013, 23 (2), 63-78
64 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
preference for explaining and predicting events on the basis of such relationships.
Cultures that differ in cognitive style (analytic vs. holistic cognition) also differ in
social orientation. Western societies tend to be more independent, whereas East Asian
societies tend to be more interdependent (Nisbett et al., 2001). The link between
cognitive style and social orientation has been widely accepted (e.g., Markus &
Kitayama, 1991b; Nisbett et al., 2001). It has also been reported that cognitive
differences and differences in social orientation between Westerners and East Asians
also hold true for differences in probability judgments and decision making (Choi,
Choi, & Norenzayan, 2004).
In recent years, the cognitive processes involved in probability judgments and
decision making have been investigated within the framework of cultural psychology
(for a review, see Yates, Lee, Sieck, Choi, & Price, 2002). Although differences in
probability judgments and decision making that have been demonstrated varies across
cultures, research is lacking in several related areas. In particular, studies in cultural
psychology have not examined ambiguity aversion (Ellsberg, 1961), which refers to
the finding that people prefer options involving known probabilities (risk) to options
involving ambiguous probabilities (ambiguity). Our present concern is whether or not
the comparative ignorance hypothesis (Fox & Tversky, 1995) is common across
cultures. According to this hypothesis, ambiguity aversion is seen when participants
evaluate lotteries with both clear and ambiguous probabilities, whereas ambiguity
aversion decreases, or disappears when they evaluate each lottery in isolation.
Therefore, the present study first examined whether the effects of ambiguity are
common across cultures. If they are common, the second problem is whether the
comparative ignorance hypothesis can explain ambiguity aversion. We used French
participants as samples of Westerners, and Japanese participants as samples of
Easterners. Moreover, this study makes a distinction between self and other choice, in
examining whether it is possible to support the comparative ignorance hypothesis by
comparing French and Japanese participants. We attempted to enlarge the comparative
ignorance hypothesis to include cross-cultural research on choice preference. We
focus on the questions of how analytic versus holistic models of cognition can
influence the processes of decision making.
Ambiguity Aversion, Competence, and Comparative Ignorance
One of the well-known paradigms that highlights of the difference between risk and
ambiguity is Ellsberg’s two-color problem (e.g., Ellsberg, 1961). Ellsberg’s two-color
problem involves two boxes, which contain red and black chips. Box 1 contains 50
red and 50 black chips, whereas Box 2 contains 100 red and black chips in a
proportion unknown to the participant. People draw a chip from one of the boxes and
win the drawing if they get a red chip. When asked which box they prefer to draw a
chip from, most people don’t express their preference for colors, however they prefer
to bet on the 50:50 probability box (Box 1) rather than on the box with the unknown
probability (Box 2). In fact, the probability of drawing a chip of either color is
identical in the two boxes. This preference for betting on known probabilities rather
than unknown probabilities is called ambiguity aversion (Ellsberg, 1961).
Einhorn and Hogarth (1985) pointed out the difference in the nature of the
uncertainty between Box 1 and Box 2. In Box 1, one is at least certain about the
uncertainty in the box. On the other hand, in Box 2, although one’s best estimate of
the probability may be .5, confidence in that estimate is low. The importance of
CULTURE, AMBIGUITY AND CHOICE 65
Ellsberg’s two-color problem lies in the difference in the nature of the uncertainty
between Box 1 and Box 2. Thus, Box 1 is favored over Box 2 because the degree of
uncertainty is known for Box 1, and the probability of success is judged to be higher
than that for Box 2, which is more ambiguous. These effects can be explained by the
competence hypothesis of Heath and Tversky (1991). Heath and Tversky (1991; Fox
& Tversky, 1995) suggested that decisions made under uncertainty depend not only
on the degree of uncertainty but also on its source. Participants preferred betting on
their beliefs in situations where they felt particularly competent or knowledgeable,
although they preferred betting on chance when they did not.
In addition to the competence hypothesis, Fox and Tversky (1995) proposed the
more progressive comparative ignorance hypothesis. They argued that the production
of ambiguity aversion in Ellsberg’s two-color problem depends on a direct comparison
of Box 1, in which the degree of uncertainty is certain, and Box 2, in which the
ambiguity provides unreliable information. They also pointed out that prior research
on ambiguity aversion had manipulated uncertainty as a within-subjects variable
operating under conditions of certainty and ambiguity. In other words, Fox and Tversky
(1995) suggested that research had looked only at situations in which a direct
comparison of conditions is made, when the participants are provided with complete
information and uncertain information. They hypothesized that if conditions of certainty
and conditions of ambiguity were used as between-subjects variables, ambiguity
aversion would decrease or disappear. In order to test the comparative ignorance
hypothesis, Fox and Tversky (1995) designed an experiment using Ellsberg's two-
color problem (Study 1). To investigate uncertainty, they created two groups: one
group made decisions while directly comparing a certain box and an ambiguous box
(variability within the participants), and the other group made decisions about the
boxes individually (variability between the participants). The results revealed
ambiguity aversion in the former (24.34 dollars for the certain box, 14.85 dollars for
the ambiguous box), and no difference in the latter (17.94 dollars for the certain box,
18.42 dollars for the ambiguous box), therefore supporting the comparative ignorance
hypothesis. Chow and Sarin (2001) and Fox and Weber (2002) obtained similar
results.1 Fox and Weber (2002) described the comparative ignorance hypothesis as
follows:
According to their comparative ignorance hypothesis, ambiguity aversion
is driven by a comparison with more familiar sources of uncertainty or
more knowledgeable people (which makes the notion of competence
more salient), and is not as pronounced in the absence of such a
comparison (where the notion of competence is less salient). (pp.
478-479)
This comparative ignorance hypothesis weighs whether people making probability
judgments feel knowledgeable about their circumstances or competence in completing
the given tasks.
1 In these studies, ambiguity aversion was observed both in comparative conditions and in non-
comparative conditions. The results in these studies support the weak version of the comparative
ignorance hypothesis. The weak version of the comparative ignorance is a reduced extent of ambiguity
aversion in the non-comparative conditions relative to the comparative conditions. Fox and Weber (2002)
note that the presence of ambiguity aversion in non-comparative contexts does not necessarily contradict
the comparative ignorance hypothesis.
66 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
The Provision of Choice and the Individual’s Sense of Personal Control
Before testing the comparative ignorance hypothesis, we need to incorporate another
variable. Decisions made under uncertainty are also influenced by the factor providing
choice. However, Fox and Tversky's (1995) experiment using Ellsberg's two-color
problem failed to take cultural differences in choice preference between Westerners
and East Asians into consideration.
Psychological research on the provision of choice usually makes the assumption
that choice is always beneficial and desirable. Some psychologists have postulated
that the provision of choice or control will enhance intrinsic motivation (e.g., Deci,
1981; Deci & Ryan, 1985; Langer, 1975). Being offered a choice, even an illusionary
or unimportant choice, has been shown to have strong motivational consequences
(e.g., Langer, 1975; Langer & Rodin, 1976). Likewise, it has also been shown that the
provision of choice can enhance an individual's sense of personal control (Taylor,
1989; Taylor & Brown, 1988).
However, several recent studies have shown that this preference for more choice is
not universal. Cross-cultural research on providing personal choice or control indicates
that East Asians are less confident than Americans about their personal ability to
control the environment. For example, Heine and Lehman (1995) showed that Japanese
students were less confident than Canadians in their personal ability to control
situations. Ji, Peng, and Nisbett (2000) showed that a relationship between perceived
personal control and task performance was observed among American, but not among
Asian, participants. Iyengar and Lepper (1999) found that personal choice enhanced
motivation more for European American than for Asian American participants. These
findings suggest that a preference for one’s own choice may be specific to Western
people.
This hypothesis is reinforced by the suggestion of Markus and Kitayama (1991a, b)
that a relationship between providing choice or control and intrinsic motivation might
not be universally applicable. From Markus and Kitayama’s (1991a, b) perspective,
the self can have two construals: independent or interdependent. For those who
possess an independent self, making a choice provides an occasion to show one's
preferences, to promote one's autonomy, and to express one’s uniqueness. In contrast,
making choices may be much less important for members of more interdependent
cultures of Asia and elsewhere. Use of personal choice may even be threatening to
individuals belonging to interdependent cultures when the outcome of personal choice
goes against the preference of the reference group.
Fox and Tversky's (1995) experiment using Ellsberg's two-color problem fails to
take these differences into account. The task instructions specified that "you" (the
participant) must decide on the color of a chip and then pull it out conditions that
enhance the individual's sense of personal control for members of an independent
culture, but possibly not for those from interdependent cultures. Fox and Tversky's
(1995) experiment did not include a “no-choice (lack of personal choice)” condition
that provides a choice to someone other than the participants. Lack of choice would
unnerve one’s confidence for Westerners than for East Asians (e.g., Heine & Lehman,
1995; Ji et al., 2000; Markus & Kitayama, 1991b). Our research is designed as a
cross-cultural study of Westerners (French participants) and East Asians (Japanese
participants). In order to enlarge the comparative ignorance hypothesis proposed by
Fox and Tversky (1995) to include cross-cultural research on choice preference, the
CULTURE, AMBIGUITY AND CHOICE 67
present study made a distinction between a choice made by the self and by another
person. Then, it examined whether comparing French and Japanese participants
would support the hypothesis. In the no-choice (hereafter, the “competitor”) condition,
an unknown other (a competitor) made the choice for the participants. We expected
that the competitor would be perceived as a usurper of an individual’s right to choose
more for French than for Japanese participants. Those of two conditions (self and
other choice) replicated the basic design of the studies that have demonstrated effects
of choice (e.g., Deci & Ryan, 1985; Langer, 1975).
OVERVIEW OF THE RESEARCH
One of the purposes of this study was to test ambiguity aversion among French and
Japanese participants. If ambiguity aversion was observed, the validity of the
comparative ignorance hypothesis proposed by Fox and Tversky (1995) was to be
tested across these two cultures. Finally, we examined whether making probability
judgments was affected by availability or by lack of choice.
General Method
This study followed the method of Fox and Tversky (1995), using Ellsberg's two-
color problem to investigate the amounts French and Japanese participants were
willing to bet on boxes in which the degree of uncertainty is certain or ambiguous.
The probability of success was 1/2 in the certain condition and unstated in the
ambiguous condition. In Experiment 1 (within-subjects design), each participant
evaluated lotteries with both clear and ambiguous probabilities. In Experiment 2
(between-subjects design), different participants evaluated each lottery in isolation.
While this research design examined personal choice in the same way as Fox and
Tversky's (1995) first experiment, the present study also introduced a competitor
choice condition in which someone else drew the chip.
Hypotheses
Emergence of ambiguity aversion. If the comparative ignorance hypothesis is
common across cultures, ambiguity aversion should be produced in comparative
betting conditions (Experiment 1). On the other hand, ambiguity aversion should
decrease or disappear in non-comparative betting conditions (Experiment 2).
Cross-cultural differences on the value of choice. We predicted that the French
participants, who are assumed to have a more independent self and a more analytic
cognition, would be more likely than the Japanese to bet larger sums of money in the
personal choice conditions (Iyengar & Lepper, 1999; Ji et al., 2000). On the other
hand, the Japanese participants, assumed to have a more interdependent self and a
more holistic cognition, would be more likely than the French to bet larger sums of
money in the competitor choice conditions (Ji et al., 2000; Markus & Kitayama,
1991b).
Choice effects within the culture. The French and Japanese participants would be
more likely to place higher bets under personal choice conditions than under competitor
choice conditions (Iyengar & Lepper, 1999; Ji et al., 2000; Langer, 1975; Taylor,
1989; Taylor & Brown, 1988). However, the difference between the personal choice
and competitor choice conditions among French participants would likely be larger
than among Japanese participants.
68 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
EXPERIMENT 1
The purpose of this experiment was to examine whether ambiguity aversion was
driven by a comparison between the certain box and the ambiguous box (variability
within the participants). We presented both the certain box and the ambiguous box to
French and Japanese participants. The second purpose of current experiment was to
examine whether cultural differences in choice preference demonstrated in previous
studies (e.g., Ji et al., 2000; Markus & Kitayama, 1991b) was observed between
French and Japanese participants and within each cultural group.
Method
Participants. The French participants consisted of one hundred and twenty
undergraduate students majoring in psychology in their third year of study at the
Université Lyon 2. The Japanese participants consisted of one hundred and fourteen
undergraduate students, enrolled at Kobe University, Kobe College, and Setsunan
University and majoring in psychology, literature, business administration, or
technology. The French participants’ mean age was 21.5 years, and the Japanese
participants’ mean age was 20.3 years.
Materials. French and Japanese versions of Fox and Tversky’s (1995) use of the
Ellsberg’s two-color problem were prepared. The problems of the personal choice
conditions are shown as follows2 3
Imagine that there is a box on the table (Box A) filled with exactly 50
red poker chips and 50 black poker chips, and a second box (Box B)
filled with 100 poker chips that are red and black, but you don’t know
their relative proportion. Suppose that you are offered a ticket to a game
that is to be played as follows:
First, you are to guess a color (red or black). Next, without looking, you
are to draw a poker chip out of one of the boxes. If the color that you
draw is the same as the one you predicted, then you will win 100 euros
(10,000 yen for the Japanese version); otherwise you win nothing. What
is the most that you would pay for a ticket to play such a game for each
of the boxes? (From 0 euros to 100 euros in the French version, and
from 0 yen to 10,000 yen in the Japanese version.)
The most that I would be willing to pay for a ticket to Box A (50 red; 50
black) is:
The most that I would be willing to pay for a ticket to Box B ( ? red; ?
black) is:
The problems of the competitor choice conditions are shown as follows:
Imagine that there is a box on the table (Box A) filled with exactly 50
red poker chips and 50 black poker chips, and a second box (Box B)
filled with 100 poker chips that are red and black, but you and a
2 All materials in Experiment 1 and Experiment 2 were first written in English, translated into French
and Japanese, and then back-translated into English. All discrepancies were resolved and the texts were
checked once again.
3 The underlined words show the difference between the personal choice conditions and the competitor
choice conditions.
CULTURE, AMBIGUITY AND CHOICE 69
competitor don’t know their relative proportion. Suppose that you are
offered a ticket to a game that is to be played as follows:
First, the competitor is to guess a color (red or black). Next without
looking, this competitor is to draw a poker chip out of one the boxes. If
the color that this competitor draws is the same as the one she or he
predicted, then you will win nothing; otherwise you win 100 euros
(10,000 yen for the Japanese version). What is the most that you would
pay for a ticket to play such a game for each of the boxes? (From 0
euros to 100 euros in the French version, and from 0 yen to 10,000 yen
in the Japanese version.)
The most that I would be willing to pay for a ticket to Box A (50 red; 50
black) is:
The most that I would be willing to pay for a ticket to Box B ( ? red; ?
black) is:
Design. The design was a 2 (nationality: French, Japanese) by 2 (choice: personal
choice, competitor choice) by 2 (degree of uncertainty: certain, ambiguity) mixed-
model design, with the first and the second factors between the participants.
Procedure. The experiment was performed in regular classes in the respective
universities. The French and Japanese participants were randomly assigned to two
groups. In the personal choice group, 61 French participants and 57 Japanese
participants themselves decided the target color of the chip to be drawn. In the
competitor choice group, with 59 French participants and 57 Japanese participants,
the target color was decided by someone else and that person drew the chip. Each
participant was given a problem that included instructions describing the two tasks on
the same page. One task described a box in which the degree of uncertainty was
certain (hereafter, the “certainty condition”), and the second task described a box in
which the degree of uncertainty was ambiguous (hereafter, the “ambiguity condition”).
The order in which the boxes were presented was counterbalanced. The questions in
the personal choice and the competitor choice conditions are stated above. The
participants evaluated the most that they would be willing to pay for a ticket to the
clear bet and the ambiguous bet from 0 euros to 100 euros in the French version, and
from 0 yen to 10,000 yen in the Japanese version. The amount of money that could be
bet corresponded to the currency and range used by Fox and Tversky (1995), 0 U.S.
dollars to 100 U.S. dollars. The economic disparity between the U.S., France, and
Japan is not large enough to influence the probability judgments.
Results and Discussion
The average amounts that participants were willing to pay under each condition are
shown in Table 1. An analysis of variance was carried out on the three factor design.4
5 The main effect of degree of uncertainty was significant (M = 28.27 vs. 15.94), F(1,
4 The analysis was carried out on 1/100 of the Japanese yen value.
5 In this study, we emphasized ambiguity aversion and culture over gender differences. Therefore, we
did not include gender in our analysis. Moreover, there were a few cells with small sample sizes for
French males. However, recent studies have demonstrated gender differences in both ambiguity aversion
and personal choice in decision making. For example, Borghans, Golsteyn, Heckman and Meijers (2009)
have showed that ambiguity aversion in Ellsberg’s two-color problem was stronger for women than for
70 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
230) = 64.98, p < .001, η2 = .06; indicating that ambiguity aversion emerged. The
main effects of nationality (M = 21.53 vs. 22.72) and choice (M = 19.53 vs. 24.71)
were not significant, F < 1, ns, η2 = .001, and F(1, 230) = 3.38, p = .07, η2 = .01,
respectively.
The two-way interaction of nationality and choice was significant, F(1, 230) = 4.13,
p < .05, η2 = .01. The two-way interactions of nationality and degree of uncertainty,
as well as choice and degree of uncertainty, were not significant, F(1, 230) = 1.62, ns,
η2 = .002, and F < 1, ns, η2 = .000, respectively. The three-way interaction of
nationality, choice, and degree of uncertainty was not significant, F(1, 230) = 3.21, p
= .07, η2 = .003.
An analysis of simple main effects from the two-way interaction between
nationality and choice revealed that the French placed higher bets than the Japanese in
the personal choice condition (M = 21.80 vs. 17.27), F(1, 230) = 4.40, p < .05, r
= .11; but lower bets in the competitor choice condition (M = 21.26 vs. 28.17), F(1,
230) = 10.27, p < .01, r = .26. The Japanese in the competitor choice group placed
higher bets than the Japanese in the personal choice group (M = 28.17 vs. 17.27), F(1,
230) = 25.52, p < .01, r = .63. However, no significant difference was found between
the amounts the French bet under the personal and competitor choice conditions (M =
21.80 vs. 21.26), F < 1, ns, r = .002.
Table 1
Results of Comparative Bets in Experiment 1
Certain box
Ambiguous box
Values
SD
Values
SD
French
group
Personal
choice
Competitor
choice
28.00
29.29
26.30
27.75
15.59
13.22
22.59
17.22
Japanese
group
Personal
choice
Competitor
choice
24.26
31.53
24.10
29.55
10.27
24.81
11.06
29.54
28.27
27.18
15.94
21.93
Note. The amount of money bet ranged from 0 euros to 100 euros n the French version,
and from 0 yen to 10,000 yen in the Japanese version. The analysis was carried out
on 1/100 of the Japanese yen value.
In summary, ambiguity aversion appeared for both French and Japanese
participants. This indicates that all participants preferred betting on known
probabilities to unknown probabilities. As expected, the French participants placed
higher bets than the Japanese participants under personal choice conditions, whereas
the Japanese participants placed significantly higher bets than the French participants
men. Ji et al. (2000) and Yamaguchi et al. (2005) have reported that personal choice is less important for
both American and Japanese women, as compared to men, regardless of their cultural context. Despite the
fact that there were less than 10 French male participants in the competitor choice condition, we
performed an analysis that included gender as one factor. A four-factor analysis of variance revealed that
main effect of gender was not significant, F < 1, ns, η2 = .000, (16 males, 89 females, 15 sex specification
omissions for French, 38 males and 76 females for Japanese; M = 24.27 vs. 21.53).
CULTURE, AMBIGUITY AND CHOICE 71
in the competitor choice condition. These findings are consistent with cultural
differences in the value of choice between Westerners and East Asians. For those who
possess a more independent self, personal choice may be deeply implicated in the
sense of self-identity (Iyengar & Lepper, 1999; Ji et al., 2000). On the other hand, use
of personal choice may pose a threat to individuals belonging to interdependent
cultures (Ji et al., 2000; Markus & Kitayama, 1991b).
However, inconsistent with our prediction, Japanese participants placed significantly
higher bets under competitor choice conditions than personal choice conditions.
Although we predicted that the French participants would be more likely to place
higher bets under personal choice conditions than competitor choice conditions, no
significant difference was found between the amounts they bet under these two
conditions. Within the Western cultural sphere, the French did not put as much weight
on personal choice as did Americans in Fox and Tversky’s (1995) study (e.g., Choi, et
al., 2004; Yates, et al., 2002). These unexpected findings regarding Japanese and
French participants are discussed below in the general discussion section.
EXPERIMENT 2
In Experiment 2, conditions of certainty and ambiguity were used as between-subjects
variables, in order to examine whether ambiguity aversion occurs even in these
conditions. Consequently, we examined whether the comparative ignorance hypothesis
is common across cultures or varies across cultures. If the comparative ignorance
hypothesis is common across cultures, ambiguity aversion should decrease or disappear
in non-comparative betting conditions. The second purpose of this experiment was to
examine whether cultural differences in choice preference demonstrated in previous
studies (e.g., Ji et al., 2000; Markus & Kitayama, 1991b) was observed between
French and Japanese participants and within each cultural group.
Method
Participants. The French participants consisted of 174 undergraduate students
majoring in psychology from the Université Lyon 2. The Japanese participants
consisted of 168 students, enrolled at Kobe University, Kobe College, and Setsunan
University majoring in psychology, literature, business administration, or technology.
The French participants’ mean age was 21.3 years, and the Japanese participants’
mean age was 19.9 years.
Materials. French and Japanese versions of Fox and Tversky’s (1995) use of the
Ellsberg’s two-color problem were prepared. However, only one of the two boxes
was presented at a time to the participants. Four different problems are shown in the
Appendix.
Design. The design was a 2 (nationality: French, Japanese) by 2 (choice: personal
choice, competitor choice) by 2 (degree of uncertainty: certain, ambiguity) design
between participants.
Procedure. The procedure was basically identical to that in Experiment 1; however,
in this case, only one of the two boxes was presented to the each participant. The
French and Japanese participants were randomly assigned to four groups. In the
personal choice group, 86 French participants (42 in the certain condition and 44 in
the ambiguous condition) and 84 Japanese participants (42 in the certain condition
72 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
and 42 in the ambiguous condition) themselves decided the target color of the chip to
be drawn. In the competitor choice group, with 88 French participants (44 in the
certain condition and 44 in the ambiguous condition) and 84 Japanese participants (42
in the certain condition and 42 in the ambiguous condition), the target color was
decided by someone else and that person drew the chip. They were given a problem
and instructions that described only one of the two tasks, either one in which the
degree of uncertainty was certain or one in which the degree of uncertainty was
ambiguous. These four different kinds of problems are shown in the Appendix.
Results and Discussion
The average amounts participants were willing to pay under each condition are shown
in Table 2. A three-factor analysis of variance revealed only a significant main effect
of uncertainty (M = 23.40 vs. 14.23), F(1, 334) = 12.08, p < .001, η2 = .04. All other
main effects and interactions were not significant, all Fs < 1, ns, all η2 < .004.6
Contrary to the predictions derived from the comparative ignorance hypothesis,
ambiguity aversion did not disappear in this experiment.
Table 2
Result of Noncomparative Bets in Experiment 2
Certain box
Ambiguous box
Values
SD
Values
SD
French
group
Personal
choice
Competitor
choice
N=42
N=44
20.45
26.82
26.54
28.98
N=44
N=44
16.07
12.50
23.05
17.07
Japanese
group
Personal
choice
Competitor
choice
N=42
N=42
22.98
23.20
20.63
29.60
N=42
N=42
14.57
13.76
20.47
16.25
23.40
26.88
14.23
19.53
Note. The amount of money bet ranged from 0 euros to 100 euros n the French version,
and from 0 yen to 10,000 yen in the Japanese version. The analysis was carried out
on 1/100 of the Japanese yen value.
Briefly stated, ambiguity aversion emerged when the French and Japanese
participants compared the two boxes in Experiment 1 and when they made decisions
concerning each box separately in Experiment 2. These results indicate that the
comparative ignorance hypothesis was not valid across French and Japanese participants.7
6 In this study, we emphasized ambiguity aversion and culture over gender differences. Therefore, we
did not include gender in our analysis. Moreover, there were a few cells with small sample sizes for
French males. Despite the fact that there were less than 10 French male participants in the competitor
choice condition, we performed an analysis that included gender as one factor. A four-factor analysis of
variance revealed that main effect of gender was not significant, F < 1, ns, η2 = .000, (21 males, 119
females, 34 sex specification omissions for French, 64 males and 104 females for Japanese; M = 20.01 vs.
19.13).
7 The degree of uncertainty in Experiment 1 was a within-subjects variable; in Experiment 2, it was a
between-subjects variable. In the former, the error within individuals can be excluded. In the latter, where
such exclusion is impossible, significant differences are more easily obtained, and this must be considered
a potential research design flaw.
CULTURE, AMBIGUITY AND CHOICE 73
Our current study shows that ambiguity aversion differs in participants from Western
culturesAmericans in Fox and Tversky’s (1995) study in particular. It is unclear
why we failed to observe a cross-cultural difference in the value of choice between
French and Japanese participants and no differences in the value of choice within
each cultural group.
GENERAL DISCUSSION
Is the Comparative Ignorance Hypothesis Common across Cultures?
Ambiguity aversion was observed both in comparative betting conditions (Experiment
1) and in non-comparative betting conditions (Experiment 2). Consequently, the
comparative ignorance hypothesis cannot explain ambiguity aversion for French and
Japanese participants. However, the comparative ignorance hypothesis should not be
completely rejected. Tests of the sizes of the uncertainty main effect across Experiments
1 and 2 revealed a smaller effect size in Experiment 2 (η2 = .04) than in Experiment 1
(η2 = .06). The comparative ignorance hypothesis may therefore have some validity.
C. R. Fox (personal communication, September, 22, 2007) asserts that there is a
strong version of the comparative ignorance effect (e.g., Fox & Tversky, 1995),
which entails a total absence of ambiguity aversion in non-comparative conditions,
and a weak version (e.g., Chow & Sarin, 2001; Fox & Weber, 2002) that entails
ambiguity aversion in non-comparative conditions, which is less extreme than in the
comparative conditions. The weak version of the comparative ignorance is a reduced
extent of ambiguity aversion in the non-comparative conditions relative to the
comparative conditions. Chow and Sarin (2001) found that the strong result obtained
by Fox and Tversky’s (1995) comparative ignorance hypothesis was more fragile and
the complete disappearance of ambiguity aversion in non-comparative condition
might not be as robust as Fox and Tversky had supposed. We thus suggest that the
results of this study support the weak version of the comparative ignorance effect
(Chow & Sarin, 2001; Fox & Weber, 2002).
Alternative interpretation. We would like to submit an alternative interpretation,
from the perspective of cultural psychology. In Experiment 2, the Japanese
participants under conditions of ambiguity seem to have been comparing the box in
which the degree of uncertainty is ambiguous with the other one in which the degree
of uncertainty is certain. We infer that probability judgments the Japanese participants
did about the ambiguous box have implications for holistic cognition of the field and
the object. Some studies (e.g., Ji, et al., 2000; Masuda & Nisbett, 2001) have been
reported that East Asians pay more attention to the field as a whole than to the focal
object and are more attuned to the relationship between the object and the field than
Westerners. In contrast, Westerners have analytic cognition that involves detachment
of the object from its context, a tendency to focus on attributes of the object to assign
the object to categories, and a preference for using rules about the categories to
explain and predict the object’s behavior (e.g., Ji, et al., 2000; Masuda & Nisbett,
2001). The Japanese participants may compare betting on unknown probabilities (i.e.,
the focal object) with known probabilities (i.e., the field) when the ambiguous box
was presented to them. Because of holistic cognition, it may have been difficult for
our Japanese participants to avoid comparing the two boxes. Unfortunately, there is
no conclusive evidence in the present study whether or not Japanese participants
actually compared the certain box and the ambiguous box. However, they may have
74 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
been more likely than the Westerners in Fox and Tversky's (1995) study to spontaneously
make a mental comparison to the box with some degree of uncertainty.
Yet why did the French participants, who share with Americans the same sphere of
Western culture, exhibit ambiguity aversion when they did not directly compare the
two types of box? These French findings in the present study offer evidence in
support of unexpected or counterintuitive cross-cultural findings in decision judgments
(Choi, et al., 2004). These results suggest that differing thought styles exist even
within a single cultural sphere (Choi, et al., 2004; Nisbett et al., 2001; Schwartz, 2000).
Choice in Probability Judgments
Cross-cultural differences in the value of choice. In Experiment 1, the French placed
higher bets than the Japanese under the personal choice conditions, and the Japanese
placed higher bets than the French under the competitor choice conditions. Some
cross-cultural studies have been reported that Westerners possess a sense of personal
control to a greater degree than East Asians (e.g., Heine & Lehman, 1995; Iyengar &
Lepper, 1999; Ji et al., 2000; Yamaguchi, Gelfand, Ohashi & Zemba, 2005). This
study revealed differences between the French and Japanese ways of feeling personal
control. Therefore, the differences in probability judgments and decision making
found here were consistent with those of previous studies (e.g., Heine & Lehman,
1995; Ji et al., 2000).
Disappearance of choice effects. In Experiment 1, as for the French participants,
we predicted that they would be more likely to place higher bets under personal
choice conditions than under competitor choice conditions (Iyengar & Lepper, 1999;
Ji et al., 2000; Langer, 1975; Taylor, 1989; Taylor & Brown, 1988). However, no
significant differences were found in the amounts the French bet under the personal
and competitor choice conditions. This finding for French participants may suggest
that within the Western cultural sphere, the French do not place as great an importance
on personal choice as did Americans in Fox and Tversky’s (1995) study.
Alternatively, it might be valid, particularly for Americans, that the provision of
choice enhances an individual's sense of personal control. Iyengar and Lepper (1999)
described the value of choice in American culture as follows:
Americans cherish choice. “Liberty,” after all, is enshrined, subordinate
only to life itself in our Declaration of Independence. Even today, the
provision and the rhetorical appeal of choice permeates [sic] American
lifefrom the plethora of options available in our grocery stores, where
there is often an entire aisle devoted solely to potato chips or to soft
drinks, to the use of the label pro-choice by abortion advocates as a
persuasive device in current political debate. (p. 349)
Secondly, in Experiment 2, we failed to observe a cross-cultural difference in
choice, between French and Japanese participants, or within each cultural group. It is
unclear as to why the choice effects disappeared in the non-comparative context. One
possible explanation is that these results, in particular related to French participants,
may have been the result of a debiasing strategy of the illusion of personal choice,
resulting from the experimental manipulation of non-comparative context.
Again unexpectedly, in Experiment 1, the Japanese participants bet more money
under the competitor choice conditions than under the personal choice conditions. It is
unclear as to why Japanese participants preferred a choice made by competitors, than
CULTURE, AMBIGUITY AND CHOICE 75
by themselves. In Japan, many people may believe it is more probable that leaving a
decision up to an unknown other (the competitor) brings them luck. Japanese
participants, who value interpersonal harmony, may prefer choices provided
indirectly, as in the competitor choice conditions, rather than direct choices (Markus &
Kitayama, 1991b). We expected that the competitor would be perceived as a usurper
of the individual’s right to choose. In this sense, similar to previous studies, it was
reasonable that this condition was considered to be a “no-choice (lack of personal
choice)” condition (e.g., Heine & Lehman, 1995; Iyengar & Lepper, 1999; Ji et al.,
2000; Markus & Kitayama, 1991b). The results of this study suggest the need to
reconsider whether the dichotomy between personal and competitor choice truly
reflects cultural differences in choice preferences, as claimed by previous studies (e.g.,
Heine & Lehman, 1995; Iyengar & Lepper, 1999; Ji et al., 2000; Markus & Kitayama,
1991b). It is also suggested that more research is needed to evaluate the adequacy of
our manipulation.
Manipulation of choice conditions. The present study didn’t introduce another lack
of personal choice conditions in which someone considered by participants to be
worthy of their trust and close in relationship drew the chip. In fact, some research
showed that the provision of personal choice was more important to Westerners (with
independent selves) than to East Asians (with interdependent selves); however,
having choices made by relevant in-group others was more important to East Asians
than to Westerners (e.g., Iyengar & Lepper, 1999; Yamaguchi et al., 2005).
We expected that participants would regard the "competitor" mentioned in the
instructions under the competitor choice conditions as a person belonging to the
out-group and that leaving the choice to someone who belongs to the out-group would
reduce participants’ confidence. The present experiments did not address the possibility
that East Asians prefer a choice made by a person trusted authority figures or peers
(belonging to the in-group) to that made themselves or by a person belonging to the
out-group (e.g., Iyengar & Lepper, 1999). One of the reasons for this could be that in
this study, we emphasized ambiguity aversion and culture over the provision of
choice made by people belonging to the out-group vs. the in-group. Further research
should examine whether making probability judgments is affected by having choices
made by relevant in-group others and would focus on the questions how analytic
versus holistic models of cognition can influence the processes of decision making.
REFERENCES
Borghans, L., Golsteyn, B., Heckman, J. J., & Meijers, H. (2009). Gender Differences
in Risk Aversion and Ambiguity Aversion. Journal of the European Economic
Association, 7, 649-658.
Choi, I., Choi, J. A., & Norenzayan, A. (2004). Culture and Decisions. In D. J.
Koehler & N. Harvey (Eds.), Blackwell Handbook of Judgment and Decision
Making (pp. 504-524). Blackwell Publishing.
Chow, C. C., & Sarin, R. K. (2001). Comparative ignorance and the Ellsberg Paradox.
Journal of Risk and Uncertainity, 22, 129-139.
Deci, E. L. (1981). The psychology of self-determination. Lexington, MA: Health.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in
human behavior. New York: Plenum.
76 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
Einhorn, H. J., & Hogarth, R. M. (1985). Ambiguity and uncertainty in probabilistic
inference. Psychological Review, 92, 433-461.
Ellsberg, D. (1961). Risk, ambiguity, and the Savage axioms. Quarterly of Journal of
Economics, 75, 643-669.
Fox, C. R. & Tversky, A. (1995). Ambiguity aversion and comparative ignorance.
Quarterly Journal of Economics, 110, 585-603.
Fox, C. R. & Weber, M., (2002). Ambiguity aversion, comparative ignorance, and
decision context. Organizational Behavior and Human Decision Processes, 88,
476-498.
Heath, C., & Tversky, A. (1991). Preference and belief: Ambiguity and competence
in choice under uncertainty. Journal of Risk and Uncertainty, 4, 5-28.
Heine, S. J., & Lehman, D. R. (1995). Cultural variation in unrealistic optimism:
Does the West feel more invulnerable than the East? Journal of Personality and
Social Psychology, 68, 595-607.
Iyengar, S.S., & Lepper, M.R. (1999). Rethinking the value of choice: A cultural
perspective on intrinsic motivation. Journal of Personality and Social Psy-
chology, 76, 349-366.
Ji, L., Peng, K., & Nisbett R. E. (2000). Culture, control, and perception of
relationships in the environment. Journal of Personality and Social Psychology,
78, 943-955.
Langer, E. J. (1975). The illusion of control. Journal of Personality and Social
Psychology, 32, 311-328.
Langer, E. J., & Rodin, J. (1976). The effects of choice and enhanced personal
responsibility for the aged: A field experiment in an institutional setting. Journal
of Personality and Social Psychology, 34, 191-198.
Markus, H. R., & Kitayama, S. (1991a). Cultural variation in the self concept. In J.
Strauss & G. R. Goethals (Eds.), The self: Interdisciplinary approaches (pp.
18-24). New York: Springer-Verlag.
Markus, H. R., & Kitayama, S. (1991b). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98, 224-253.
Masuda, T., & Nisbett R. E. (2001). Attending holistically versus analytically:
Comparing the context sensitivity of Japanese and Americans. Journal of
Personality and Social Psychology, 81, 922-934.
Nisbett, R. E. (2003). The Geography of Thought: How Asians and Westerners Think
Differently…and Why. New York: The Free Press.
Nisbett, R. E., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of
thought: Holistic vs. analytic cognition. Psychological Review, 108, 291-310.
Schwartz, B. (2000). Self-determination: The tyranny of freedom, American
Psychologist, 55, 79-88.
Taylor, S. E. (1989). Positive illusions: Creative self-deception and the healthy mind.
New York: Basic Books.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological
perspective on mental health. Psychological Bulletin, 103, 193-210.
Yamaguchi, S., Gelfand, M., Ohashi, M. M., & Zemba, Y. (2005). The cultural
psychology of control: Illusions of personal versus collective control in the
United States and Japan. Journal of Cross-Cultural Psychology, 36, 750-761.
Yates, J. F., Lee, J. W., Sieck, W. R., Choi, I., & Price, P. C. (2002). Probability
Judgment across Cultures. In T. Gilovich, and D. Kahneman (eds.), Heuristics
CULTURE, AMBIGUITY AND CHOICE 77
and Biases (pp. 271-291). Cambridge: Cambridge University Press.
Key words: Comparative ignorance hypothesis, Ambiguity aversion, Probability
judgments, Choice, Cultural difference
78 ADACHI, YAMA, HENST, MERCIER, KARASAWA, KAWASAKI
APPENDIX: PROBLEMS USED IN EXPERIMENT 2
In the personal choice group the certain condition. Imagine that there is a box on
the table filled with exactly 50 red poker chips and 50 black poker chips. Suppose that
you are offered a ticket to a game that is to be played as follows:
First, you are to guess a color (red or black), Next, without looking, you are to draw a
poker chip from that box. If the color that you draw is the same as the one you
predicted, then you will win 100 euros (10,000 yen for the Japanese version);
otherwise you win nothing. What is the most that you would pay for a ticket to play
such a game for the box? (From 0 euros to 100 euros in the French version, and from
0 yen to 10,000 yen in the Japanese version.)
The most that I would be willing to pay for a ticket to the box is:
In the competitor choice group the certain condition. Imagine that there is a box on
the table filled with exactly 50 red poker chips and 50 black poker chips. Suppose that
you are offered a ticket to a game that is to be played as follows:
First, a competitor is to guess a color (red or black). Next without looking, this
competitor is to draw a poker chip from that box. If the color that this competitor
draws is the same as the one she or he predicted, then you will win nothing; otherwise
you win 100 euros (10,000 yen for the Japanese version). What is the most that you
would pay for a ticket to play such a game for the box? (From 0 euros to 100 euros in
the French version, and from 0 yen to 10,000 yen in the Japanese version.)
The most that I would be willing to pay for a ticket to the box is:
Note. The underlined words are to show the difference between the personal choice
group and the competitor choice group in the certain conditions.
In the personal choice group the ambiguity condition. Imagine that there is a box on
the table filled with 100 poker chips that are red and black, but you don’t know their
relative proportion. Suppose that you are offered a ticket to a game that is to be
played as follows:
First, you are to guess a color (red or black). Next, without looking, you are to draw a
poker chip from that box. If the color that you draw is the same as the one you
predicted, then you will win 100 euros (10,000 yen for the Japanese version);
otherwise you win nothing. What is the most that you would pay for a ticket to play
such a game for the box? (From 0 euros to 100 euros in the French version, and from
0 yen to 10,000 yen in the Japanese version.)
The most that I would be willing to pay for a ticket to the box is:
In the competitor choice group the ambiguity condition. Imagine that there is a box
on the filled with 100 poker chips that are red and black, but you and a competitor
don’t know their relative proportion. Suppose that you are offered a ticket to a game
that is to be played as follows:
First, the competitor is to guess a color (red or black). Next without looking, this
competitor is to draw a poker chip from that box. If the color that this competitor
draws is the same as the one she or he predicted, then you will win nothing; otherwise
you win 100 euros (10,000 yen for the Japanese version). What is the most that you
would pay for a ticket to play such a game for the box? (From 0 euros to 100 euros in
the French version, and from 0 yen to 10,000 yen in the Japanese version.)
The most that I would be willing to pay for a ticket to the box is:
Note. The underlined words are to show the difference between the personal choice
group and the competitor choice group in the ambiguity conditions.
Article
To improve the provision of microcredit to farmers, this paper describes and suggests which cognitive biases in Behavioral Economics may apply to and should be tested on borrowers (farmers) and lenders (workers at Rural Credit Cooperatives) in China’s rural market for microcredit. For example, some of the biases considered are loss aversion, ambiguity aversion and the sequential bias. If biases are found, this paper also suggests what would help might mitigate the biases’ negative effects. This paper is the first to apply Behavioral Economics to the lenders in China’s rural credit market.
Full-text available
Article
This study focused on the cultural psychology of control in the United States and Japan. The authors tested a hypothesis that Japanesewould tend to overestimate their ability to control their outcomes collectively compared to personally, whereas Americans would show the reverse tendency. As expected, Japanese participants in the group condition, relative to those in the individual condition, were more optimistic about obtaining a favorable outcome. American men, on the other hand, were more optimistic in the individual condition. Interestingly, similar to the Japanese participants, American women showed a reverse but nonsignificant tendency to be more optimistic in the group condition. These results indicate that the psychology of control is both gendered and cultured.
Article
Americans now live in a time and a place in which freedom and autonomy are valued above all else and in which expanded opportunities for self-determination are regarded as a sign of the psychological well-being of individuals and the moral well-being of the culture. This article argues that freedom, autonomy, and self-determination can become excessive, and that when that happens, freedom can be experienced as a kind of tyranny. The article further argues that unduly influenced by the ideology of economics and rational-choice theory, modern American society has created an excess of freedom, with resulting increases in people's dissatisfaction with their lives and in clinical depression. One significant task for a future psychology of optimal functioning is to deemphasize individual freedom and to determine which cultural constraints are necessary for people to live meaningful and satisfying lives.
Book
I: Background.- 1. An Introduction.- 2. Conceptualizations of Intrinsic Motivation and Self-Determination.- II: Self-Determination Theory.- 3. Cognitive Evaluation Theory: Perceived Causality and Perceived Competence.- 4. Cognitive Evaluation Theory: Interpersonal Communication and Intrapersonal Regulation.- 5. Toward an Organismic Integration Theory: Motivation and Development.- 6. Causality Orientations Theory: Personality Influences on Motivation.- III: Alternative Approaches.- 7. Operant and Attributional Theories.- 8. Information-Processing Theories.- IV: Applications and Implications.- 9. Education.- 10. Psychotherapy.- 11. Work.- 12. Sports.- References.- Author Index.
Chapter
As American society prepares to participate in the internationalized world of the 21st century, the first reports suggested there was little to fear. Perhaps it was a small world after all. Preferences for Coke, McDonald’s, rock and rap music, copying machines, and T-shirts appear universal. Along with these confirmations of cross-cultural similarity in consumer behavior, however, come anecdotes that hint at some powerful differences. And this seems particularly true for the differences between the West and the East. The popular press recount the examples daily. While Americans learn the value of standing out and being noticed and hold to the maxim that “it’s the squeaky wheel that gets the grease,” a prevalent proverb in the East—in Japan, China, even in Australia—is “the nail that sticks up shall be hammered down.” Nowhere are East–West differences more apparent than in baseball. Both cultures love the game. And, in fact, Japanese baseball is painstakingly modeled on American baseball. Yet in Japan, the goal is not the familiar one of obliterating the opposing team by scoring as many runs as possible, but instead, to beat the other team by a face-savingly small amount. Moreover, rather than applauding the virtues of “doing your own thing,” Japanese coaches caution that “lone wolves are the cancer of the team.” While American players are advised to express themselves and “get it off their chest,” among Japanese players such behavior is considered immature and untoward—the goal is to keep one’s emotions to oneself (see Whiting, 1989, for a full description of these differences).