Culture as a Moderator of Epistemically Suspect Beliefs

Abstract

A consistent finding reported in the literature is that epistemically suspect beliefs (e.g., paranormal beliefs) are less frequently endorsed by individuals with a greater tendency to think analytically. However, these results have been observed predominantly in Western participants. In the present work, we explore various individual differences known to predict epistemically suspect beliefs across both Western and Eastern cultures. Across four studies with Japanese (n = 666) and Western (n = 650) individuals, we find that the association between thinking style and beliefs varied as a function of culture. Specifically, while Westerners who scored higher on measures of Type-2 analytic thinking tended to endorse epistemically suspect beliefs less, this association was not observed in Japanese samples, suggesting that the often-observed negative association between analytic thinking and epistemically suspect beliefs may be exclusive to Western individuals. Additionally, we demonstrate that a tendency to think holistically (specifically with regards to causality) is positively associated with the endorsement of epistemically suspect beliefs within both samples. Overall, we discuss how various individual differences predict the endorsement of epistemically suspect beliefs across cultures.

Full text

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 1
ORIGINAL RESEARCH
published: 09 February 2022
doi: 10.3389/fpsyg.2022.745580
Edited by:
Jean Baratgin,
Université Paris 8, France
Reviewed by:
Jose Yong,
Northumbria University,
United Kingdom
Keiko Ishii,
Nagoya University, Japan
*Correspondence:
Yoshimasa Majima
majima.y@hokusei.ac.jp
Specialty section:
This article was submitted to
Cognition,
a section of the journal
Frontiers in Psychology
Received: 22 July 2021
Accepted: 03 January 2022
Published: 09 February 2022
Citation:
Majima Y, Walker AC, Turpin MH
and Fugelsang JA (2022) Culture as
a Moderator of Epistemically Suspect
Beliefs. Front. Psychol. 13:745580.
doi: 10.3389/fpsyg.2022.745580
Culture as a Moderator of
Epistemically Suspect Beliefs
Yoshimasa Majima1*, Alexander C. Walker2, Martin Harry Turpin2and
Jonathan A. Fugelsang2
1Department of Psychology for Well-Being, Hokusei Gakuen University, Sapporo, Japan, 2Department of Psychology,
University of Waterloo, Waterloo, ON, Canada
A consistent finding reported in the literature is that epistemically suspect beliefs (e.g.,
paranormal beliefs) are less frequently endorsed by individuals with a greater tendency
to think analytically. However, these results have been observed predominantly in
Western participants. In the present work, we explore various individual differences
known to predict epistemically suspect beliefs across both Western and Eastern
cultures. Across four studies with Japanese (n= 666) and Western (n= 650)
individuals, we find that the association between thinking style and beliefs varied as
a function of culture. Specifically, while Westerners who scored higher on measures
of Type-2 analytic thinking tended to endorse epistemically suspect beliefs less, this
association was not observed in Japanese samples, suggesting that the often-observed
negative association between analytic thinking and epistemically suspect beliefs may
be exclusive to Western individuals. Additionally, we demonstrate that a tendency to
think holistically (specifically with regards to causality) is positively associated with the
endorsement of epistemically suspect beliefs within both samples. Overall, we discuss
how various individual differences predict the endorsement of epistemically suspect
beliefs across cultures.
Keywords: epistemically suspect beliefs, cultural differences, analytic thinking, analytic-holistic cognition,
bullshit receptivity
INTRODUCTION
Fundamentally, scientists are in the business of trying to improve the accuracy of both their
own and humanity’s beliefs via the collection of information about the universe. As such, it is
no surprise that a community of researchers have become profoundly interested in epistemically
suspect beliefs (ESBs), which refer to beliefs that do not cohere with established scientific evidence
(e.g., paranormal beliefs; Lobato et al., 2014). Previous studies investigating ESBs have focused on
the individual differences of believers as opposed to skeptics, including differences in cognitive
ability, reasoning skills, and thinking style. These findings suggest that those who endorse ESBs
are in general less educated (Gray and Mill, 1990;Aarnio and Lindeman, 2005), perform worse on
some reasoning tasks (Blackmore and Tro´
scianko, 1985;Roberts and Seager, 1999), and engage less
in analytic thinking compared to skeptics (Lindeman and Aarnio, 2006;Pennycook et al., 2012).
Analytic thinking in this context refers to thought processes that are commonly characterized as
being deliberative, reflective, and requiring working memory (i.e., Type-2 processes; Evans, 2008;
Evans and Stanovich, 2013). Much research has shown that analytic thinking is a good predictor of
Frontiers in Psychology | www.frontiersin.org 1February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 2
Majima et al. Culture and Epistemically Suspect Beliefs
performance on a range of reasoning and decision making
tasks (e.g., heuristics-and-biases tasks that involve successfully
overriding an incorrect intuitive response; Toplak et al., 2011).
With regard to the endorsement of everyday irrational beliefs,
such as superstitions, research has suggested that these beliefs
arise from our intuitive processes (i.e., Type-1 processes; Evans,
2008;Evans and Stanovich, 2013), tend to rely on heuristics, seek
for coherent causal explanations, and favor evidence providing
supports for one’s current beliefs (Risen, 2016). Therefore, a
line of argumentation like ESBs providing simple surface-level
explanations about the universe may be intuitively appealing.
These beliefs will be maintained unless they are more closely
re-examined by analytic thinking. In line with this claim, other
work has found that the tendency to engage in analytic thinking
is negatively associated with various forms of ESBs including
religious belief (Gervais and Norenzayan, 2012;Pennycook et al.,
2012, 2014), belief in the paranormal (Aarnio and Lindeman,
2005;Lobato et al., 2014), and pseudoscientific beliefs (Lindeman,
2011;Lobato et al., 2014). Relatedly, analytic thinking has
been shown to be negatively associated with receptivity to
superficially impressive yet vacuous statements (i.e., pseudo-
profound bullshit; Pennycook et al., 2015a;Walker et al., 2019).
On the basis of these findings, many scholars have argued
that ESBs are rooted in Type-1 intuitive processing which
can be overridden by effortful and Type-2 analytic processes
(Pennycook et al., 2015b).
Although the negative association between ESBs and analytic
thinking appears robust, it may be reasonable to be skeptical
about the underlying mechanisms proposed. First, in the domain
of religious belief, studies have shown contradictory findings. For
example, the role of analytic thinking in supporting religious
disbelief has been challenged by research showing that promoting
analytical thinking does not promote religious disbelief (Yonker
et al., 2016;Sanchez et al., 2017). Furthermore, most studies
examining the association between cognitive style and ESBs have
been conducted exclusively with WEIRD (Western, Educated,
Industrialized, Rich and Democratic; Henrich et al., 2010)
participants. Therefore, it remains an open question whether
the link between analytic thinking and ESBs generalizes to
non-WEIRD populations. It is important to examine possible
cultural differences in the underlying processes associated with
ESBs in order to better understand everyday irrational beliefs.
If cultural differences in thinking styles are identified between
populations, they are likely to interact with interventions focused
on thinking (e.g., education, and debiasing). Furthermore, cross-
cultural comparisons of everyday beliefs are important as they
can lead to a better understanding of cultural differences
on the effects of various psycho-social factors on well-being
(Yong et al., 2021).
Recent work has begun to investigate cross-cultural differences
as they relate to ESBs. For example, compared to Westerners,
Chinese individuals have been found to be more likely
to endorse paranormal beliefs (Shiah et al., 2010) and
Turkish individuals more likely to endorse conspiracy beliefs
(Bruder et al., 2013). Relatedly, Bahçekapili and Yilmaz (2017)
reported a series of studies featuring Muslim populations
showing that analytic thinking was negatively associated with
intrinsic/extrinsic motivations for religiosity (e.g., personal duty
or societal pressure), but positively associated with religiosity
dealing with an open-minded seeking of answers to existential
questions. Based on these findings, Bahçekapili and Yilmaz
(2017) suggest that the link between analytic thinking and
religiosity depends on how religiosity is expressed among
individuals. Relatedly, Tosyali and Aktas (2021) show that the
negative link between analytic thinking and superstitious beliefs
is stronger for Turkish participants with low-to-moderate levels
of religiosity than highly religious individuals. These results
suggest that the relationship between analytic thinking and
irrational beliefs is not as simple as analytic thinking always
suppressing such beliefs, but may involve group differences
related to norms surrounding belief, such as culture. In the
domain of paranormal and pseudoscientific belief, Japanese
individuals self-reporting a strong tendency toward analytic
thinking, measured by the rationality subscale of the Rational-
Experiential Inventory (Pacini and Epstein, 1999), were more
likely to hold paranormal and pseudoscientific beliefs (Karasawa
and Tsukimoto, 2010;Majima, 2015), providing initial evidence
that the commonly observed negative association between
analytic thinking and ESBs may not generalize to non-WEIRD
populations. Nevertheless, participants’ level of analytic thinking
was self-reported within this study, leaving open the possibility
that they were simply mis-calibrated in their self-assessment.
Conversely, it could be that the link between analytic thinking
and ESBs is absent in Japanese samples, perhaps on account
of ESBs being less in violation of Japanese as opposed to
Western cultural norms.
Along with differences in ESBs, findings from cultural
psychology have demonstrated differences in cognitive style
between Western and Eastern populations. These findings
suggest that Westerners are more likely to adopt ‘analytic’ modes
of cognition, while Easterners are more likely to take holistic
approaches (Nisbett et al., 2001;Nisbett and Miyamoto, 2005;
see also Buchtel and Norenzayan, 2009 for discussions regarding
differences between Type-2 analytic thinking as described by
contemporary dual process theorists and ‘analytic’ cognition
as described in the domain of cultural psychology). Generally
speaking, in the domain of cultural psychology, ‘analytic’
individuals tend to focus on the specific attributes or elements
of an object or problem, rather than the larger context as a whole.
In contrast, rather than focusing on individual elements, holistic
individuals tend to focus on the totality of an object or problem,
including the overarching context. Furthermore, holistic thinking
has been shown to predict the acceptance of mutually conflicting
statements (naïve dialecticism; Spencer-Rodgers et al., 2010) and
complex causality (Maddux and Yuki, 2006). Relatedly, past work
demonstrates that Eastern (i.e., Japanese) participants tend to
report more mixed emotions than Americans, predominantly in
pleasant situations (Miyamoto et al., 2010).
These findings suggest that differences in the endorsement of
ESBs across Western and Eastern cultures may be explained, at
least partially, by cultural differences in analytic-holistic modes
of thinking. Since holistic cognition has a more dialectical
Frontiers in Psychology | www.frontiersin.org 2February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 3
Majima et al. Culture and Epistemically Suspect Beliefs
orientation, such as the acceptance of contradiction and complex
causation, it is reasonable to assume that holistic thinkers may
be more likely to accept mutually conflicting statements and as a
result show greater endorsement of ESBs than those engaging in
more rule-based, or ‘analytic’ modes of reasoning. Therefore, the
tendency for Eastern individuals to think holistically may offer
one explanation for why Eastern individuals seemingly endorse
more ESBs compared to Western individuals. However, readers
may by puzzled by the distinction between Type-2 analytic
thinking, as discussed by dual process theorists, and an ‘analytic’
mode of cognition, as discussed by cross-cultural psychologists.
For this difference, Buchtel and Norenzayan (2009) argue that
Type-2 analytic thinking and cultural ‘analytic’ cognition share
similarities with regards to context independence and weak
attention to social relations. Nevertheless, holistic cognition
does not necessarily correspond to Type-1 processes, nor does
‘analytic’ cognition necessarily correspond to Type-2 processes.
Rather, Buchtel and Norenzayan (2009) posit that analytic-
holistic modes of thinking can be viewed as different styles,
or individual variations of thinking, that operate under the
umbrella of Type-2 processes. From this perspective, the negative
association between so-called Type-2 analytic thinking and ESBs
may not necessarily be culturally universal.
The present research investigates endorsement of ESBs
(e.g., paranormal and pseudoscientific beliefs) within samples
of Western (North American and Western European) and
Eastern (Japanese) participants. We seek to not only assess the
frequency of ESBs across Western and Eastern populations,
but also investigate whether such beliefs are predicted by
the same individual difference variables (e.g., Type-2 analytic
thinking) across cultures. Based on past work (Shiah et al.,
2010;Bruder et al., 2013), we expect Western participants
to endorse ESBs less than Eastern participants. We examine
whether such a difference can be explained by Western
participants preferring more ‘analytic’ and linear styles of
thinking compared to Eastern participants, who more frequently
engage in holistic and dialectic styles of thinking (Nisbett et al.,
2001;Nisbett and Miyamoto, 2005).
Lastly, we investigate receptiveness to pseudo-profound
bullshit across Western and Eastern cultures. Similar to
holding ESBs, endorsement of pseudo-profound bullshit (i.e.,
statements that are superficially impressive yet consist of
a largely random assortment of profound-sounding words)
has been argued to result from a failure to engage Type-
2 analytic thinking (Pennycook et al., 2015a;Pennycook and
Rand, 2020). That is, endorsement of pseudo-profound bullshit
shares a common cognitive mechanism with acceptance of
ESBs. Relatedly, studies have demonstrated positive associations
between bullshit receptivity and real-world beliefs, such as the
endorsement of paranormal beliefs (Pennycook et al., 2015a)
and “fake news” (Pennycook and Rand, 2020). Nevertheless,
the claim that bullshit receptivity naturally follows from a
failure to engage Type-2 analytic thinking may be incomplete
as other factors such as the tendency to perceive patterns
or maintain radically subjective beliefs may similarly explain
receptiveness to pseudo-profound bullshit statements (Turpin
et al., 2019;Walker et al., 2019). These factors likely vary
based on culture and so too may receptiveness to pseudo-
profound bullshit.
ETHICS STATEMENT
All studies were conducted in accordance with APA ethical
standards and approved by the relevant ethics committees.
All individuals gave their informed consent online prior
to participation.
STUDY 1
Method
Participants
A sample of 298 participants were recruited from two online
crowdsourcing platforms, 147 Japanese participants (59% female;
Mage = 38.37, SDage = 9.45) from CrowdWorks (CW) and
151 North American and European participants (47% female;
Mage = 33.23, SDage = 11.50; 36% United States residents, 38%
United Kingdom residents, 25% other) from Prolific Academic
(ProA)1. All participants received compensation (CW = 240 JPY;
ProA = £2.00) upon completion of an online questionnaire. For
all studies, participants were required to possess an approval
rating of 95% or higher on either CW or ProA in order to
be eligible to participate. We collected our full sample prior to
data analyses, report all data exclusions, all manipulations, and
all measures used.
Materials
Paranormal Belief Scale
We assessed the degree to which participants endorsed various
paranormal beliefs by asking them to judge the plausibility of 12
paranormal belief items (BPA). These items were drawn from the
Psi questionnaire (6-items; Roberts and Seager, 1999; Japanese
version adopted from Majima, 2015) and Revised Paranormal
Belief Scale (6-items; adopted from Tobacyk, 20042). Participants
judged the plausibility of each BPA item (e.g., “How likely is it that
you possess some form of ‘psychic ability’?”) using a five-point
scale ranging from 1 (Extremely Unlikely) to 5 (Certain). Ratings
1We did not conduct an a priori power analysis to determine sample size for
this study. Instead, decisions regarding sample size were made with reference to
related past work (e.g., Pennycook et al., 2012;Majima, 2015). We also decided to
include Prolific participants having non-European origin into our Western sample
since our preliminary analysis revealed that the overall results were not different
when we excluded these participants. The number of corresponding participants
were 6, 13, 4 and 51 in Studies 1 through 4, respectively. Similarly, CrowdWorks
participants with a non-Asian origin were labeled as Japanese participants, since
preliminary analysis also revealed that the results were not different when we
excluded non-Asian CrowdWorks participants. The number of non-Asian CW
participants were 0, 10, 1 and 24 in Studies 1 through 4, respectively. Some
studies suggest that people of Asian ethnicity living in Western countries provide
responses resembling an intermediate between Asian people living in Eastern
countries and European people living in Western countries (Norenzayan et al.,
2002). Therefore, it is noteworthy that the present results (featuring our full
sample) did not differ when excluding Japanese participants with a non-Asian
origin and Western participants with an Asian origin.
2The Japanese translations for R-PBS items were drawn from two previous works
(Nakajima et al., 1992, 1993).
Frontiers in Psychology | www.frontiersin.org 3February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 4
Majima et al. Culture and Epistemically Suspect Beliefs
given to each item were averaged to calculate a BPA score for each
participant. A complete list of items (for all studies) can be viewed
in the supplementary materials.
Pseudoscientific Belief Scale
We assessed the degree to which participants endorsed
pseudoscientific beliefs with 12 pseudoscientific belief (BPS) items
(e.g., “Homeopathic remedies foster spontaneous healing”). This
scale consisted of six items from Majima (2015), three items from
Lobato et al. (2014), and three items from Dekker et al. (2012)3.
Participants were asked to judge their agreement with each item
on a five-point scale ranging from 1 (Disagree) to 5 (Agree). Once
again, ratings given to each item were averaged to obtain a BPS
score for each participant.
Cognitive Style Measures
We assessed participants’ tendency to engage in rational
(analytic) and experiential (intuitive) thinking with 12 items
drawn from Naito et al.’s (2004) Information-Processing Style
Inventory (IPS). The original IPS contains 24 items adapted
from Pacini and Epstein (1999) Rational-Experiential Inventory
(REI), consisting of six items from each of the four REI subscales
(rational engagement, rational ability, experiential engagement,
and experiential ability). In order to reduce participants’ work
load, we chose 12 items (three items from each subscale) showing
high factor loading scores in studies with Western (Pacini and
Epstein, 1999) and Japanese (Naito et al., 2004) participants. For
each item, participants were presented with a statement and asked
to judge the extent to which the statement was true of themselves
on a five-point scale (1 = Definitely not true, 5 = Definitely
true). Responses to six rationality items (e.g., “I enjoy intellectual
challenges”) and six experiential items (e.g., “I like to rely on my
intuitive impressions”) were averaged to obtain a rationality and
experientiality score for each participant.
Participants were also presented with a three item Cognitive
Reflection Test (CRT; Frederick, 2005), designed to provide a
behavioral measure of participants’ ability to suppress an intuitive
incorrect response in favor of an analytical correct answer. For
all CRT items participants provided their answers in a free-entry
text box. These two scales were introduced to measure individual
differences in Type-2 analytic (rationality and CRT) and Type-1
intuitive (experientiality) thinking.
Lastly, we administered the Analysis-Holism scale (AHS;
Choi et al., 2007) to assess participants’ tendencies to engage
in analytic-holistic modes of thinking. The original AHS scale
consisted of 24 items evenly split into four subscales (causality,
attitude toward contradiction, perceived change, and locus
of attention). However, to reduce participants’ work load,
we only administered 12 AHS items (3 items from each
subscale). Participants were presented each item individually and
responded using a seven-point scale ranging from 1 (Strongly
disagree) to 7 (Strongly agree). Within each subscale, ratings given
to each item were averaged to obtain an AHS subscale score.
3Translations for BPS and AHS items were created as following. First, the first
author translated items into Japanese. Next, another scholar working as an English
teacher at the same university of the first author translated items back into English.
The final translation was decided through discussion between the above two
people.
Cognitive Ability Measures
We assessed participants’ cognitive ability with a syllogistic
reasoning (cf. Majima, 2015) and numeracy task. We measured
participants’ logical reasoning ability using a syllogistic reasoning
task adopted from Markovits and Nantel (1989). The syllogistic
reasoning task presented participants with eight syllogisms, all
of which featured a conflict between the logical validity of the
syllogism and the believability of its conclusion. Importantly, this
design ensured that for each item the intuitive response (based
on conclusion believability) would need to be overridden in
order for the correct answer to be produced. For each syllogism,
participants were asked to indicate whether the conclusion
followed logically from the premises presented, irrespective
of the believability of the concluding sentence. The sum of
correctly solved syllogisms was used as an index of the logical
reasoning ability of each participant. Additionally, we measured
participants’ numeracy using the Subjective Numeracy Scale
(SNS; Fagerlin et al., 2007). This scale was developed and
validated as a conventional self-evaluation numeracy scale. We
computed an unweighted mean of Percent of Maximum Possible
(POMP; Cohen et al., 1999) scores for both our syllogistic
reasoning and numeracy task in order to obtain a joint measure
of cognitive ability (CAB).
Procedure
Participants were administered an online questionnaire in which
they were asked to complete several tasks in the following
order: pseudoscientific belief items, paranormal belief items,
information-processing style inventory items (we refer to this
scale as the REI hereafter referencing the origin of these
items), a syllogistic reasoning task, the SNS, the AHS, and the
CRT. Following completion of these tasks participants were
asked a series of demographic questions (i.e., age, gender,
nationality, ethnicity, native language, and highest educational
level). Among these questions, only age and gender were
considered predictors for the subsequent analysis4. All materials
were presented in Japanese for CW participants and in English
for ProA participants.
Results and Discussion
Table 1 displays the descriptive statistics and bivariate
correlations for all key variables. With the exception of
AHS subscales (αranged from .60 to .75)5, all scales showed
good internal consistency (αranged from .79 to .94). A series of
independent samples t-tests found that our two samples did not
differ in cognitive ability, CRT performance, or experientiality
scores (all ps>.10). However, Westerners scored higher on
our measure of rationality (i.e., the six rationality subscale
items included in the REI), t(296) = 6.22, p<.001, d= 0.72.
Consistent with past work, Japanese participants endorsed
4Responses to nationality, ethnicity and native language questions were originally
included as indicators to detect participants from a different culture than the one in
which they were currently living. However, as noted in text footnote 1, we decided
not to exclude participants on this basis. Therefore, we did not use these variables
in the subsequent analysis.
5Note that the reliability scores for AHS subscales were similar to those found in
the original study (αranged from .56 to .71; Choi et al., 2007).
Frontiers in Psychology | www.frontiersin.org 4February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 5
Majima et al. Culture and Epistemically Suspect Beliefs
TABLE 1 | Descriptive statistics and correlation of key variables (Study 1).
West (n= 151) JP (n= 147) Sample difference Pearson correlational coefficients a
αM SD M SD t p d 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
1. Age 33.23 11.50 38.37 9.45 4.22 <.001 0.49 −.003 .023 −.001 .029 .014 .030 .081 −.044 −.010 .199
2. CAB 66.62 21.10 62.95 18.04 1.61 .108 0.19 .160 .305 −.274 .490 .018 .015 −.040 .002 −.215 −.165
3. RAT 0.87 3.80 0.79 3.20 0.88 6.22 <.001 0.72 −.063 .453 −.259 .177 .071 .102 .011 .235 −.084 −.106
4. EXP 0.88 3.18 0.93 3.04 0.77 1.44 .151 0.17 .033 −.285 −.125 −.299 .029 −.201 −.139 −.147 .481 .372
5. CRT 1.63 1.18 1.44 1.09 1.47 .143 0.17 .098 .646 .342 −.335 .033 −.048 .097 .051 −.255 −.273
6. Cause 0.75 4.89 1.37 5.33 1.06 −3.09 .002 0.36 .054 −.151 −.066 .221 −.169 .053 −.057 .195 .081 .133
7. Contra 0.66 4.98 1.22 4.84 1.12 1.02 .307 0.12 −.031 −.104 −.068 .178 −.137 .015 −.035 .206 −.079 .121
8. Change 0.60 4.74 1.14 5.15 1.18 −3.00 .003 0.35 −.017 −.066 .016 .019 .079 −.162 .003 .014 −.199 −.161
9. Attention 0.64 4.51 1.30 5.28 0.91 −5.92 <.001 0.69 .001 .017 −.062 .024 −.052 .028 .097 −.075 −.045 −.069
10. BPA 0.94 2.24 1.03 2.87 0.83 −5.81 <.001 0.66 .126 −.389 −.192 .393 −.486 .283 .112 −.121 −.042 .534
11. BPS 0.79 2.53 0.68 2.87 0.53 −4.81 <.001 0.56 .183 −.339 −.211 .291 −.454 .353 .110 −.171 −.031 .674
aValues above the diagonal indicate coefficients for Japanese sample, and values below the diagonal indicates coefficients for Western sample. The coefficients shown in bold face were significant at p <.05.
d, Cohen’s d statistics; α, Cronbach’s alpha coefficients; CAB, cognitive ability score described by unweighted mean POMP scores (0–100) of syllogism and subjective numeracy; RAT, rationality (five-point); EXP,
experientiality (five-point); CRT, Cognitive Reflection Test; Cause, Causality scale of AHS (seven-point); Contra, Attitude toward Contradiction scale of AHS; Change, Perception of Change scale of AHS; Attention, Locus
of Attention of AHS; BPA, belief in the paranormal (five-point); BPS, belief in pseudoscience (five-point). Of the above indicators, RAT and CRT are indices of Type-2 analytic thinking, whereas EXP is an index of intuitive
thinking. All AHS subscales are indices of a holistic mode of thought.
more paranormal, t(296) = 5.81, p<.001, d= 0.66, and
pseudoscientific beliefs, t(296) = 4.81, p<.001, d= 0.56, than
Westerners. They also scored higher in three of the four AHS
subscales: causality [t(296) = 3.09, p= .002, d= 0.36], perception
of change [t(296) = 3.00, p= .003, d= 0.35], and locus of
attention [t(290) = 5.92, p<.001, d= 0.69]. Surprisingly, no
difference between samples was observed for the attitude toward
contradiction AHS subscale, t(296) = 1.02, p= .307, d= 0.12.
Cultural Differences and Determinants of
Epistemically Suspect Beliefs
In order to identify potential determinants of ESBs we conducted
multiple regression analyses predicting ESBs (i.e., paranormal
beliefs and pseudoscientific beliefs; see Table 2). In order to
identify whether effects of our predictors differed across cultures
(i.e., were moderated by cultural affiliation), we adopted a
hierarchical regression approach. We excluded cognitive ability
and three subscales of the AHS (attitude toward contradiction,
perception of change, and locus of attention) from these analyses
because preliminary analyses failed to show any contribution
of these variables to differences in ESBs. In the first step, all
predictors were entered simultaneously into the model. Next, the
interaction terms of sample and other predictors (i.e., rationality,
experientiality, CRT, and causality) were entered. For paranormal
beliefs, 1R2s demonstrated a significant improvement in the
prediction of paranormal beliefs by introducing interaction terms
of sample and other predictors (1R2= .02, p= .005). We also
found significant improvements in predicting pseudoscientific
beliefs (1R2= .01, p= .043). The final models for both
paranormal and pseudoscientific beliefs can be viewed in
Table 2 (left panel).
We observed a significant gender difference for both ESBs
in which women were more likely to have stronger ESBs
compared to men (BPA:β=−.45, p<.001, BPS:β=−.23,
p= .027). As predicted, CRT performance was negatively
associated with ESBs (BPA:β=−.36, p<.001, BPS:β=−.38,
p<.001), whereas a holistic understanding of causality was
positively associated with ESBs (BPA:β= .14, p= .023, BPS:
β= .24, p<.001). Additionally, we observed significant
experientiality ×sample (BPA:β= .24, p= .016, BPS:β= .21,
p= .042) and CRT ×sample interactions for both ESBs (BPA:
β= .33, p= .001, BPS:β= .26, p= .013). Subsequent simple
slope analyses revealed cultural affiliation to be a significant
moderator of the association between experientiality and ESBs.
The unstandardized simple slopes for Japanese participants
were, BPA = 0.08, p<.001, and BPS = 0.04, p<.001,
and BPA = 0.04, p= .004, and BPS = 0.01, p= .250 for
Western participants. Lastly, we found cultural affiliation to
be a significant moderator of the association between CRT
performance and ESBs. The unstandardized simple slopes
for Japanese participants were, BPA =−0.02, p= .709, and
BPS =−0.07, p= .121, and BPA =−0.31, p<.001, and
BPS =−0.21, p<.001 for Western participants. Therefore,
CRT performance predicted ESBs in our Western sample but
not Japanese sample, possibly reflecting differences in cultural
values with regards to avoiding ESBs. On the other hand, a
propensity to value intuition was a strong predictor of ESBs
Frontiers in Psychology | www.frontiersin.org 5February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 6
Majima et al. Culture and Epistemically Suspect Beliefs
TABLE 2 | Regression analyses predicting beliefs by cognitive traits, cultural orientation, and interactions with sample.
Study 1 Study 2
BPA (adj. R2= .39) BPS (adj. R2= .33) BPA (adj. R2= .22) BPS (adj. R2= .30)
Predictors bSE βp b SE βp b SE βp b SE βp
(Intercept) 2.33 0.17 <.001 2.29 0.12 <.001 2.24 0.18 <.001 2.26 0.12 <.001
Sample a0.57 0.10 .576 <.001 0.22 0.07 .343 .001 0.43 0.13 .474 .001 0.20 0.08 .318 .015
Gender b−0.45 0.10 −.454 <.001 −0.15 0.07 −.232 .027 −0.20 0.10 −.217 .041 −0.05 0.06 −.076 .450
Age 0.01 0.00 .058 .218 0.01 0.00 .187 <.001 0.00 0.01 .048 .398 0.01 0.00 .132 .015
RAT 0.00 0.01 .020 .790 −0.00 0.01 −.025 .758 −0.04 0.02 −.170 .041 −0.04 0.01 −.277 <.001
EXP 0.04 0.01 .183 .004 0.01 0.01 .077 .250 0.10 0.02 .431 <.001 0.06 0.01 .348 <.001
CRT −0.31 0.06 −.358 <.001 −0.21 0.04 −.384 <.001 −0.21 0.06 −.271 .001 −0.21 0.04 −.376 <.001
Causality 0.04 0.02 .136 .023 0.04 0.01 .238 <.001
DSS 0.29 0.12 .188 .016 0.02 0.08 .018 .804
Sample ×RAT 0.01 0.02 .052 .609 0.00 0.01 .036 .738 0.08 0.03 .350 .003 0.04 0.02 .264 .019
Sample ×EXP 0.05 0.02 .242 .016 0.03 0.01 .214 .042 −0.06 0.03 −.238 .036 −0.01 0.02 −.057 .596
Sample ×CRT 0.29 0.09 .331 .001 0.15 0.06 .264 .013 0.16 0.09 .208 .057 0.07 0.06 .132 .203
Sample ×Causality −0.03 0.03 −.102 .290 −0.02 0.02 −.128 .206
Sample ×DSS −0.14 0.19 −.089 .473 0.24 0.13 .217 .065
All continuous variables except for age were mean-centered. a1 = Japanese, 0 = Western. b1 = men, 0 = women.
among Japanese participants, whereas it was not predictive for
Western participants, particularly for pseudoscientific beliefs.
Furthermore, the fact that a propensity toward multiple causality
positively predicted pseudoscientific beliefs provides some
support for the idea that holistic modes of thought can lead
to the endorsement of ESBs. Of note however, the influence of
thinking style along the analytic-holistic dimension did not vary
across cultures.
STUDY 2
Study 1 demonstrated that holistic understanding of causality
partially explained participants’ endorsement of ESBs. However,
participants’ attitude toward contradiction was not associated
with endorsement of ESBs. This may be due to the fact that
individuals’ attitude toward contradiction in the current context
doesn’t reflect their propensity toward naïve dialecticism (i.e.,
the belief that things are changing continuously, tolerance of
contradiction, and the preference for endorsing moderate options
centered between two opposing options; cf. Zhang et al., 2015).
Therefore, in Study 2, we further explored the association
between dialectic thinking and endorsement of ESBs.
Method
Participants
A sample of 316 participants were recruited from two online
crowdsourcing platforms, 167 Japanese participants (56% female;
Mage = 39.26, SDage = 9.45) from CW and 149 Western
participants (42% female; Mage = 30.60, SDage = 10.04) from
ProA. All participants received compensation (CW = 240 JPY;
ProA = £2.00) upon completion of an approximately 16-min
online questionnaire. As gender was a predictor in our regression
analyses, two participants from ProA, and one from CW were
excluded due to not reporting gender.
Materials
The materials used in Study 2 were nearly identical to those
used in Study 1, with the following exceptions. First, we reduced
the number of REI items we administered from 12 to 10 (now
including five rational subscale and five experiential subscale
items). These items were adopted from the REI-10 (Epstein
et al., 1996), and assessed individual differences in thinking
style. Second, we no longer administered the AHS, instead
presenting participants with 32 Dialectical Self Scale items as
our measure of holistic cognition (DSS; Spencer-Rodgers et al.,
2004; the Japanese version of DSS items were adopted from
Zhang et al., 2015). This scale measured participants’ degree of
dialectical thinking (i.e., the degree to which participants are able
to synthesize competing ideas or viewpoints) within the domain
of self-perception (e.g., “When I hear two sides of an argument, I
often agree with both”). Participants responded to each DSS item
using a seven-point scale ranging from 1 (Strongly disagree) to 7
(Strongly agree). Responses to all 32 items were averaged for each
participant to obtain a DSS score.
Frontiers in Psychology | www.frontiersin.org 6February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 7
Majima et al. Culture and Epistemically Suspect Beliefs
Procedure
Participants were administered an online questionnaire in which
they were asked to respond to various items in the following
order: pseudoscientific belief items, paranormal belief items,
REI items, a syllogistic reasoning task, the SNS, the DSS, and
the CRT. Following the completion of these tasks, participants
were asked a series of demographic questions (i.e., age, gender,
ethnicity, and native language). As in Study 1, all materials were
presented in Japanese for CW participants and in English for
ProA participants.
Results and Discussion
Table 3 displays the descriptive statistics and bivariate
correlations of all key variables. A series of independent
samples t-tests demonstrated that our two samples did not differ
in cognitive ability or CRT performance (both ps>.19). In
contrast, we observed differences in our measures of paranormal
beliefs, pseudoscientific beliefs, dialectical thinking, rationality,
and experientiality between Japanese and Western samples. As
in Study 1, Japanese participants endorsed more paranormal,
t(313) = 4.42, p<.001, d= 0.50, and pseudoscientific beliefs,
t(314) = 3.97, p<.001, d= 0.45. Japanese participants also
demonstrated a greater degree of dialectic thinking compared to
Western participants, t(314) = 9.75, p<.001, d= 1.10. Lastly,
Western participants scored higher on both the rationality,
t(314) = 6.50, p<.001, d= 0.73, and experientiality, t(314) = 6.40,
p<.001, d= 0.72, subscales of the REI-10.
Cultural Differences and Determinants of
Epistemically Suspect Beliefs
As in Study 1, we conducted multiple regression analyses
predicting ESBs (i.e., paranormal beliefs and pseudoscientific
beliefs) in order to identify potential determinants of ESBs
(see Table 2). We adopted a hierarchical regression approach
to identify whether the effects of predictors were moderated
by cultural affiliation. Consistent with Study 1, we found
that women were more likely to endorse paranormal beliefs
(β=−.32, p= .015) and older individuals more likely to
endorse pseudoscientific beliefs (β=−.22, p= .041). Additionally,
experientiality was positively associated with ESBs (BPA:β= .43,
BPS:β= .35, ps<.001), while rationality (BPA:β=−.17, p= .041,
BPS:β=−.28, p<.001) and CRT performance (BPA:β=−.27,
p= .001, BPS:β=−.38, p<.001) was negatively associated with
ESBs. Lastly, DSS scores were found to be positively associated
with the endorsement of paranormal but not pseudoscientific
beliefs (BPA:β= .19, p= .016, BPS:β= .02, p= .804).
We observed multiple interactions between various cognitive
traits and cultural affiliation. For belief in the paranormal, we
found significant rationality ×sample (β= .35, p= .003) and
experientiality ×sample interactions (β=−.24, p= .036),
as well as a marginally significant CRT ×sample interaction
(β= .21, p= .057). Deconstructing the rationality ×sample
interaction, we found a negative slope within our Western
sample (unstandardized slope = −0.038, p= .040) and a positive
slope within our Japanese sample (0.045, p= .015). Similarly,
CRT performance was shown to be negatively associated with
paranormal beliefs in our Western sample (−0.212, p<.001)
TABLE 3 | Descriptive statistics and correlations of key variables (Study 2).
West (n= 149) JP (n= 167) Sample difference Pearson correlation coefficients a
αM SD M SD t p d Age CAB RAT EXP CRT DSS BPA BPS
Age 30.60 10.03 39.47 9.34 −7.92 <.001 0.92 .063 .118 −.098 .163 −.134 −.098 −.059
CAB 62.49 12.97 60.65 12.39 1.29 .199 0.15 .067 .395 −.208 .441 −.025 −.037 −.304
RAT 0.79 3.66 0.77 3.10 0.75 6.50 <.001 0.73 .105 .368 −.091 .381 −.209 .117 −.171
EXP 0.80 3.59 0.70 3.08 0.73 6.40 <.001 0.72 .127 .017 .092 −.201 −.211 .217 .336
CRT 1.41 1.15 1.39 1.19 0.10 .917 0.01 .054 .588 .236 −.015 −.057 -.064 −.329
DSS 0.83 3.62 0.59 4.20 0.45 −9.75 <.001 1.10 −.171 −.075 −.252 −.166 −.156 .019 .124
BPA 0.93 2.24 0.94 2.69 0.85 −4.42 <.001 0.50 .159 −.275 −.221 .365 −.334 .197 .468
BPS 0.79 2.48 0.69 2.76 0.57 −3.97 <.001 0.45 .202 −.317 −.281 .312 −.390 .070 .624
aValues above the diagonal indicate coefficients for Japanese sample, and values below the diagonal indicate coefficients for Western sample. The coefficients shown in bold face were significant at p <.05.
All scales except DSS (seven-point) are five-point scales. As in Study 1, RAT and CRT are indices of Type-2 analytic styles of thinking and EXP is an index of intuitive thinking. DSS is an index of holistic thinking.
DSS, Dialectical Self Scale; BPA, belief in the paranormal; BPS, belief in pseudoscience.
Frontiers in Psychology | www.frontiersin.org 7February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 8
Majima et al. Culture and Epistemically Suspect Beliefs
and not associated with such beliefs in our Japanese sample
(−0.058, p= .321). Deconstructing the experientiality ×sample
interaction, we found a positive simple slope in our Western
sample (0.104, p<.001) and a more moderate, but significant,
positive slope in our Japanese sample (0.050, p= .007). For belief
in pseudoscience, we observed significant rationality ×sample
(β= .26, p= .019) and marginally significant DSS ×sample
interactions (β= .22, p= .065). Simple slope analyses revealed
that rationality was negatively associated with belief for Western
participants (−0.044, p<.001), but not associated with belief
for Japanese participants (−0.001, p= .917). Thus, highly
rational Westerners were less likely to endorse ESBs whereas
rationality scores were not associated with ESBs in our Japanese
sample. Conversely, DSS scores were positively associated with
pseudoscientific belief in our Japanese sample (0.236, p= .015),
but not in our Western sample (0.019, p= .814), demonstrating
that Japanese participants high in dialectical thinking endorsed
more pseudoscientific beliefs while the same was not true in
our Western sample. Overall, the results of Study 2 provide
further evidence suggesting that the mechanisms underlying
ESBs differ across cultures. Specifically, rationality and Type-
2 analytic thinking, were negatively correlated with ESBs in
Westerners, but not Japanese participants. Furthermore, in some
cases, these relations were reversed (i.e., positive) within our
Japanese sample.
STUDY 3
In Studies 1 and 2, we observed that a tendency toward
holistic cognition was positively associated with ESBs. However,
we failed to find an association between propensity toward
contradiction and ESBs. As discussed earlier, the self-rated
measure of attitude toward contradiction, namely, the attitude
toward contradiction subscale of the AHS, may not reflect
an individual’s dialectical thinking behavior. Therefore, Study
3 was conducted to see whether individuals who think
dialectically have stronger paranormal and pseudoscientific
beliefs than those who do not. To this end, participants in
Study 3 were asked to indicate to what extent they endorse
statements expressing epistemically suspect beliefs (pro-belief),
as well as statements which explicitly deny epistemically suspect
beliefs (anti-belief).
Method
Participants
A sample of 301 participants were recruited from two online
sources, 151 Japanese participants (69% female; Mage = 36.62,
SDage = 8.99) from CW and 150 Western participants (43%
female; Mage = 32.91, SDage = 11.42; 17% United States
residents, 41% United Kingdom residents, 39% other) from
ProA. All participants received compensation (CW = 240 JPY;
ProA = £2.00) upon completion of an approximately 18-min
online questionnaire. One participant from the CW sample was
excluded from all analyses due to providing incomplete responses
to belief tasks.
Materials
The materials used in Study 3 were similar to those used in
Study 1, with the following exceptions. First, we administered
only 12 AHS items (those from the causality and attitude
toward contradiction subscales), removing the 12 AHS items
associated with the perception of change and locus of attention
subscales on account of responses to these items failing to explain
endorsement of ESBs. However, we decided not to exclude six
items from attitude toward contradiction subscale to see if it
correlates with actual dialectical thinking behavior. Second, we
chose five paranormal belief and five pseudoscientific belief items
from Studies 1 and 2 (e.g., “Some people can have a dream
that has predicted some future events”) and created anti-belief
statements for each item (e.g., “No one can have a dream that
has predicted future events”). The resulting 10 pairs of statements
were divided into two sets, such that each set contained five pro-
and five anti-belief items. We ensured that pro- and anti-belief
statements for the same item were not included in the same
set. As in Studies 1 and 2, participants rated their agreement
with each presented statement on a seven-point scale ranging
from “1 (Strongly disagree)” to “7 (Strongly agree).” A paranormal
(BPA) and pseudoscience (BPS) score was calculated for each
participant by taking the average of their responses to paranormal
and pseudoscientific pro-belief statements, respectively.
Lastly, we assessed Dialectic Thinking (DT) using a
methodology featured in past work (Zhang et al., 2015). In order
to calculate a DT score for each participant, believability ratings
for paired pro- and anti-belief statements were standardized
(Z-scores) for each issue such that the midpoint “4” was set
to equal zero. Next, both Z-scores were summed to create a
DT score for a particular item. Finally, all five DT scores were
averaged for paranormal (DTPA) and pseudoscientific items
(DTPS), separately, creating a DT score pertaining to paranormal
beliefs and another score pertaining to pseudoscientific beliefs.
Z-score Pk(ZPk) = (raw rating of Pro-belief statement of item
k−4)/SDPk+Ak
Z-score Ak(ZAk) = (raw rating of Anti-belief statement of item
k−4)/SDPk+Ak
DT score =1
5
5
X
k=1
|ZPk +ZAk |
Procedure
Participants were administered an online questionnaire in which
they were asked to respond to various items in the following
order: a belief task, REI-10 items, the AHS, a second belief task, a
syllogistic reasoning task, the SNS and the CRT. Following the
completion of these tasks, participants were asked a series of
demographic questions (i.e., age, gender, nationality, ethnicity,
and native language). As in Studies 1 and 2, all materials were
presented in Japanese for CW participants and in English for
ProA participants.
Results and Discussion
Table 4 shows the descriptive statistics and bivariate correlations
for all key variables. Once again, we conducted a series of
Frontiers in Psychology | www.frontiersin.org 8February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 9
Majima et al. Culture and Epistemically Suspect Beliefs
TABLE 4 | Descriptive statistics and correlations of key variables (Study 3).
West (n= 135) JP (n= 126) Sample difference Pearson correlation coefficients a
M SD M SD t p d 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
1. Age 32.91 11.42 36.62 8.99 −3.13 .002 0.36 .100 .124 −.032 .127 .077 .089 .026 .264 −.158 −.234
2. CAB 61.33 19.07 57.79 17.15 1.69 .092 0.20 −.079 .117 −.046 .454 .068 −.073 −.104 .022 −.053 −.089
3. RAT 3.69 0.71 3.15 0.79 6.18 <.001 0.72 −.118 .210 .026 .129 .208 −.120 .078 .099 −.063 −.037
4. EXP 3.65 0.72 3.32 0.67 5.29 <.001 0.61 .029 −.251 −.057 −.055 .114 .143 .226 .146 .038 .007
5. CRT 1.31 1.16 1.43 1.00 −0.99 .322 0.11 −.189 .376 .182 −.145 −.057 .062 −.305 −.096 −.143 −.059
6. Cause 5.07 0.88 5.43 0.90 −3.50 <.001 0.40 −.087 −.128 .079 .167 .075 .178 .262 .196 .046 .008
7. Contra 4.77 0.98 4.77 0.82 0.03 .978 0.00 −.053 −.111 −.222 −.024 −.179 .158 −.010 .053 −.040 .082
8. BPA 3.03 1.33 3.83 1.38 −5.15 <.001 0.60 .058 −.223 −.107 .385 −.225 .417 .070 .430 −.067 −.159
9. BPS 3.54 0.87 3.63 0.76 1.20 .229 0.14 .255 −.291 −.174 .283 −.338 .272 .161 .513 −.059 −.366
10. DTPA 0.36 0.33 0.38 0.27 −0.48 .630 0.06 .042 −.137 −.139 .224 −.048 .034 −.004 .085 .196 .164
11. DTPS 0.54 0.38 0.63 0.42 −2.00 .046 0.23 −.162 −.156 .007 .107 −.109 −.072 .087 −.087 −.046 .226
aValues above the diagonal indicate coefficients for Japanese sample, and values below the diagonal indicates coefficients for Western sample. The coefficients shown in bold face were significant at p <.05.
All scales except subscales of AHS (Cause and Contra; seven-point) are five-point scales.
DTPA, Dialectic Thinking score for paranormal items; DTPS, Dialectic Thinking score for pseudoscience items.
independent samples t-tests to investigate differences across our
Japanese and Western samples. We observed no differences
between Western and Japanese participants with regards to
their attitude toward contradiction, CRT performance, cognitive
ability, or endorsement of pseudoscientific beliefs (all ps>.09).
Replicating the findings of Studies 1 and 2, Western participants
scored higher on our measure of rationality, t(299) = 6.14,
p<.001, d= 0.72, and endorsed less paranormal beliefs,
t(299) = 5.15, p<.001, d= 0.60. Western participants also
scored higher on our measure of experientiality, t(299) = 5.29,
p<.001, d= 0.61, whereas Japanese participants scored
higher in causal perception (as measured by the causality
subscale of the AHS), t(299) = 3.50, p<.001, d= 0.40.
Interestingly, DT scores were not correlated with either causality
or contradiction scores.
Cultural Differences and Determinants of
Epistemically Suspect Beliefs
As in Studies 1 and 2, we investigated potential determinants
of ESBs by conducting hierarchical regression analyses. We
excluded measures of cognitive ability and attitude toward
contradiction as preliminary analyses failed to show any
contribution of these variables in predicting the endorsement
of ESBs. With the exception of these exclusions, all predictors
were entered simultaneously into our model in Step 1. In
Step 2, we entered the interaction terms with sample. The
introduction of these interaction terms marginally improved our
model for pseudoscientific beliefs (1R2= .02, p= .056), but
not paranormal beliefs (1R2= .01, p= .143). In Step 3, we
entered DT scores for both ESB domains (pseudoscientific and
paranormal beliefs) as well as the DT ×sample interaction
into our model. These additions improved our model for
pseudoscientific beliefs (1R2= .04, p<.001), but not
paranormal beliefs (1R2= .002, p= .243). Both Step 2
(without DT scores) and Step 3 (with DT scores) models
for both paranormal and pseudoscientific beliefs can be
viewed in Table 5.
As in Study 1, our measure of experientiality was positively
associated with ESBs (βPA = .25, βPS = .19, ps<.05) and CRT
performance was negatively associated with ESBs (βPA =−.14,
βPS =−.26, ps<.05). Additionally, holistic understanding
of causality was positively associated with ESBs (βPA = .38,
βPS = .32, ps<.001). Therefore, consistent with past theorizing,
the tendency to think holistically may leave one susceptible
to ESBs while analytic thinking may protect against such
beliefs. Unlike in Studies 1 and 2, significant interactions with
sample were found only for dialectic thinking pertaining to
pseudoscientific beliefs (β=−.24, p= .019). Simple slope
analysis revealed that dialectic thinking was positively associated
with pseudoscientific beliefs in our Western sample (0.047,
p<.001), but not our Japanese sample (0.018, p= .116). Thus,
Western participants high in dialectical thinking endorsed more
pseudoscientific beliefs while the same was not true in our
Japanese sample. Interestingly, the actual behavior of dialectic
thinking exhibited a slightly different pattern of results from self-
report measures of dialecticism. Overall, Study 3 provided further
Frontiers in Psychology | www.frontiersin.org 9February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 10
Majima et al. Culture and Epistemically Suspect Beliefs
TABLE 5 | Regression analyses predicting beliefs by cognitive traits, cultural orientation, dialectical thinking, and interactions with sample (Study3).
BPA BPS
Without DT With DT Without DT With DT
Predictors b SE βp b SE βp b SE βp b SE βp
(Intercept) 3.09 0.27 <.001 3.12 0.28 <.001 3.29 0.16 <.001 3.39 0.19 <.001
Sample a0.77 0.16 .548 <.001 1.00 0.24 .552 <.001 −0.22 0.10 −.266 .031 0.12 0.15 −.206 .438
Gender b−0.35 0.15 −.247 .022 −0.35 0.15 −.248 .022 −0.19 0.09 −.236 .041 −0.15 0.09 −.184 .106
Age 0.01 0.01 .036 .476 0.00 0.01 .028 .574 0.02 0.00 .232 <.001 0.02 0.00 .199 <.001
RAT −0.03 0.03 −.077 .328 −0.03 0.03 −.077 .330 −0.03 0.02 −.123 .142 −0.03 0.02 −.127 .120
EXP 0.10 0.03 .248 .001 0.10 0.03 .248 .001 0.04 0.02 .175 .023 0.04 0.02 .185 .015
CRT −0.19 0.09 −.144 .031 −0.19 0.09 −.146 .030 −0.19 0.05 −.249 .001 −0.20 0.05 −.263 <.001
Causality 0.10 0.02 .379 <.001 0.10 0.02 .379 <.001 0.05 0.01 .320 <.001 0.05 0.01 .316 <.001
DT c0.01 0.30 .002 .970 −0.07 0.16 −.035 .649
RAT ×sample 0.05 0.04 .137 .197 0.05 0.04 .131 .218 0.03 0.02 .169 .136 0.03 0.02 .167 .132
EXP ×sample −0.02 0.04 −.044 .671 −0.02 0.04 −.040 .701 −0.01 0.03 −.039 .722 −0.01 0.02 −.051 .639
CRT ×sample −0.17 0.13 −.131 .193 −0.19 0.13 −.146 .149 0.13 0.08 .169 .117 0.12 0.08 .161 .126
Causality ×sample −0.05 0.03 −.206 .044 −0.05 0.03 −.199 .051 −0.03 0.02 −.213 .050 −0.03 0.02 −.198 .064
DT ×sample c−0.61 0.47 −.130 .194 −0.49 0.21 −.241 .019
Adjusted R2.31 .31 .21 .25
All continuous variables except for age were mean-centered.
a1 = Japanese, 0 = Western. b1 = men, 0 = women. cDT, Dialectical Thinking score of each domain (DTPA for BPA and DTPS for BPS.
Frontiers in Psychology | www.frontiersin.org 10 February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 11
Majima et al. Culture and Epistemically Suspect Beliefs
evidence suggesting that the mechanisms underlying ESBs differ
across cultures.
STUDY 4
Study 4 explored cultural differences in receptivity to pseudo-
profound bullshit. The fact that individuals’ receptivity to
pseudo-profound bullshit is positively associated with their
acceptance of ESBs suggests common cognitive mechanisms
underlying bullshit receptivity and acceptance of ESBs. While
past work has examined the mechanisms underlying bullshit
receptivity (Pennycook et al., 2015a;Walker et al., 2019), no
work (to our knowledge) has examined these mechanisms cross-
culturally. As evidenced by Studies 1–3, the association between
various individual difference measures (e.g., analytical thinking)
and ESBs (e.g., pseudoscientific belief) may vary between
cultures. Similar to ESBs, bullshit receptivity has been found
to be negatively associated with analytical thinking (Pennycook
et al., 2015a;Pennycook and Rand, 2020), however, such studies
have relied heavily on WEIRD samples. Thus, in Study 4
we aim to investigate whether the same mechanisms (e.g.,
analytical thinking) underlie bullshit receptivity across Western
(American) and Eastern (Japanese) participants. Furthermore,
recent studies have demonstrated how individual differences in
illusory pattern perception relate to bullshit receptivity (Walker
et al., 2019) and various ESBs (van Prooijen et al., 2018).
Therefore, in Study 4, we also assessed participants’ performance
on a pattern perception task. We predict that individuals
demonstrating greater illusory pattern perception will also show a
greater receptivity to pseudo-profound statements. Additionally,
we assess whether the association between illusory pattern
perception and bullshit receptivity differs between American and
Japanese individuals.
Method
Participants
A sample of 401 participants were recruited from two online
crowdsourcing platforms, 201 Japanese participants (46% female;
Mage = 39.45, SDage = 10.02) from CW and 200 U. S.
residents (37% female; Mage = 34.30, SDage = 10.67) from
Amazon’s Mechanical Turk (MTurk). All participants received
compensation (CW = 270 JPY; Mturk = $2.50 USD) upon
completion of an approximately 15-min online questionnaire.
Materials
In Study 4, we administered the CRT and AHS (24-item version;
Choi et al., 2007) along with two new measures: a modified snowy
pictures task (Whitson and Galinsky, 2008) and a profundity
judgment task (Pennycook et al., 2015a).
Modified Snowy Pictures Task
The modified snowy pictures task (MSPT; Whitson and Galinsky,
2008) was used to measure participants’ ability to detect real
patterns and avoid endorsing illusory patterns. Participants were
presented with 24 pictures, 12 of which contained a difficult to
perceive object and 12 of which contained only visual noise.
For each picture, they were asked whether the presented image
contained an object and responded with either a “yes” or
“no” response. A non-illusory pattern perception score was
calculated for each participant by determining how many object-
present items they correctly endorsed as containing an object.
Similarly, an illusory pattern perception score was calculated for
each participant by determining how many object-absent items
participants correctly endorsed as not containing an object. Thus,
higher scores for object-present and object-absent items were
indicative of a greater tendency to identify real patterns and a
reduced tendency to endorse illusory patterns, respectively.
Profundity Judgments
Participants were presented with 30 statements (10 pseudo-
profound bullshit statements, 10 motivational quotations, and 10
mundane statements) and were asked to assess the profundity of
each statement on a five-point scale ranging from “1 = Not at all
profound” to 5 “Very profound.” All statements originated from
Pennycook et al. (2015a)6. Pseudo-profound bullshit statements
were originally obtained from two websites7able to create
superficially impressive yet meaningless statements by randomly
arranging a list of profound sounding words in a way that
maintains syntactic structure. A bullshit receptivity score (BSR)
was created for each participant by calculating their mean
profundity rating given to pseudo-profound bullshit items (e.g.,
“Hidden meaning transforms unparalleled abstract beauty”),
with higher scores indicating greater bullshit receptivity.
Ten motivational quotations and mundane statements were
also presented to participants, contrasting the meaningless
nature of pseudo-profound bullshit statements. All motivational
quotations were originally obtained via an internet search
and were designed to communicate something meaningful
and reasonably profound (e.g., “A wet man does not fear
the rain”). Mundane statements also communicated something
meaningful, however, they were designed so that the message
being communicated was banal (e.g., “New born babies
require constant attention”). Identical to that of the BSR, a
motivational quotation (MQ) and mundane statement (MS)
score was calculated for each participant by determining their
mean profundity rating given to motivational and mundane
items, respectively.
Procedure
Participants were once again administered an online
questionnaire for which they completed 30 profundity
judgments, 24 MSPT items, 24 AHS items, the CRT, and a
host of demographic questions in that order. As in previous
studies, all materials were presented in Japanese for CW
participants and in English for Mturk participants.
6The Japanese version of the profundity judgment task was created through the
following procedure. First, the first author translated items into Japanese. Next,
a professional English-proofing company translated the Japanese version of all
items back into English and the accuracy of the translation was checked against
the original text. Based on their suggestions, the first author made minor revisions
to the translation, and finally, the first and second authors decided on the final
translation by consensus.
7http://wisdomofchopra.com and http://sebpearce.com/bullshit/
Frontiers in Psychology | www.frontiersin.org 11 February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 12
Majima et al. Culture and Epistemically Suspect Beliefs
TABLE 6 | Descriptive statistics of cognitive traits and beliefs (Study 4).
West (n= 181) JP (n= 177) Sample difference Pearson correlation coefficients a
M SD M SD t p d 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
1. Age 34.30 10.67 39.45 10.02 −4.97 <.001 0.50 .165 .046 .130 .079 −.040 .075 .157 −.101 −.021 −.056
2. CRT 1.54 0.70 1.43 1.11 1.21 .227 0.12 .057 −.087 .084 .050 −.127 .151 −.087 −.049 −.160 −.400
3. Pattern (P) 9.03 2.60 8.55 2.37 1.92 .055 0.19 .084 .059 −.523 .160 −.013 −.012 .139 .155 .234 .036
4. Patten (A) 9.89 2.33 9.61 2.44 1.21 .227 0.12 .014 .165 −.376 −.054 .008 .146 −.071 −.250 −.213 −.242
5. Cause 5.18 0.92 5.36 0.73 −2.17 .030 0.22 −.035 .020 .064 .069 .139 .142 .341 .103 .207 −.026
6. Contra 4.91 1.05 4.73 0.73 2.00 .046 0.20 −.116 .066 .108 −.050 .237 −.133 .292 .024 .177 .079
7. Change 4.53 0.97 4.86 0.75 −3.83 <.001 0.38 .084 .142 .097 .225 .014 −.003 −.111 −.023 −.002 −.152
8. Attention 4.95 0.94 5.16 0.61 −2.66 .008 0.27 −.009 −.159 .010 −.004 .230 .244 −.146 .087 .167 −.006
9. BSR 2.60 1.00 2.58 0.75 0.28 .778 0.03 −.157 −.382 .116 −.334 .147 .104 −.169 .089 .527 .332
10. MQ 3.20 0.76 2.99 0.65 2.94 .003 0.29 −.033 −.178 .089 −.270 .130 .267 −.123 .193 .575 .442
11. MS 1.50 0.86 1.96 0.83 −5.52 <.001 0.55 −.112 −.358 −.096 −.453 −.062 −.027 −.516 .100 .452 .292
Pattern (P), number of correct responses to 12 object-present items where an object was embedded; Pattern (A), number of correct responses to 12 object-absent items where an object was not embedded (lower
scores illustrate greater illusory pattern perception); BSR, mean profundity rating given to pseudo-profound bullshit statements; MQ, mean profundity rating given to motivational quotes; MS, mean profundity rating
given to mundane statements. All profundity judgments are five-point scales and subscales of AHS are seven-point scales.
aValues above the diagonal indicate coefficients for Japanese sample, and values below the diagonal indicates coefficients for Western sample. The coefficients shown in bold face were significant at p <.05.
Results and Discussion
Descriptive statistics and bivariate correlations for all variables
can be viewed in Table 6. Independent samples t-tests revealed
that Japanese participants scored higher on 3 of the 4 sub-scales
of the AHS [Causality: t(399) = 2.17, p= .030, d= 0.22; Change:
t(399) = 3.83, p= .001, d= 0.38; Attention: t(399) = 2.66,
p= .008, d= 0.27] demonstrating an overall greater tendency
to engage in holistic thinking compared to United States
participants. No differences were observed between Japanese and
United States participants with regards to CRT performance,
illusory pattern perception or bullshit receptivity (all ps>.220).
Furthermore, consistent with past work (Walker et al., 2019),
bullshit receptivity was found to positively correlate with illusory
pattern perception [United States: r(198) = .33, p<.001;
Japan: r(198) = .25, p<.001]. Also consistent with past
work (Pennycook et al., 2015a;Pennycook and Rand, 2020),
CRT performance was found to negatively relate to bullshit
receptivity within our United States sample, r(198) = −.38,
p<.001. Interestingly, no association was found between CRT
performance and bullshit receptivity for Japanese participants,
r(198) = −.05, p= .487.
Next, we conducted multiple regression analyses predicting
bullshit receptivity with individual differences in CRT
performance, holistic cognition (as measured by the AHS),
illusory pattern perception, and interactions of these variables
with culture. Since preliminary analyses revealed that the AHS
subscales of attitude toward contradiction, perception of change,
and locus of attention failed to predict individuals’ receptivity to
bullshit, the causality subscale was the only measure of holistic
thinking included in our model. Analyses of 1R2showed that our
model was significantly improved at all steps (Step 1: 1R2s = .144,
p<.001; Step 2: 1R2s = .057, p<.001). The final model can be
viewed in Table 7. These results show that individual differences
in CRT performance (β=−.50, p<.001), illusory pattern
perception (β=−.34, p<.001), and holistic cognition (as
measured by the causality subscale of the AHS; β= .18, p= .002)
TABLE 7 | Regression analysis predicting bullshit receptivity by reflective thinking,
holistic cognition, illusory pattern perception, and interactions with
sample (Study 4).
Predictors b SE βp
(Intercept) 2.91 0.16 <.001
Sample a−0.05 0.08 −.048 .595
Gender b0.10 0.08 .114 .230
Age −0.01 0.00 −.109 .022
Pattern absent −0.13 0.03 −.342 <.001
CRT −0.48 0.08 −.503 <.001
Causality 0.03 0.01 .179 .002
Sample ×Pattern 0.06 0.03 .152 .100
Sample ×CRT 0.46 0.10 .479 <.001
Sample ×Causality −0.01 0.02 −.075 .425
Pattern absent, number of correct responses to object-absent items where
an object was not embedded (lower scores illustrate greater illusory
pattern perception). Adjusted R2= .182, F(9,390) = 10.90, p <.001.
a1 = Japanese, 0 = Western. b1 = men, 0 = women.
Frontiers in Psychology | www.frontiersin.org 12 February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 13
Majima et al. Culture and Epistemically Suspect Beliefs
predicted bullshit receptivity. Furthermore, we observed a
significant CRT ×sample interaction (β= .48, p<.001).
A simple slope analysis revealed that CRT was negatively
associated with bullshit receptivity within our United States
sample (−0.48, p<.001), and shared no association with bullshit
receptivity in our Japanese sample (−0.023, p= .666), suggesting
that the mechanisms underlying bullshit receptivity, like ESBs,
may differ across cultures.
GENERAL DISCUSSION
The present study investigated cultural differences related to the
endorsement of ESBs and pseudo-profound bullshit. Specifically,
we examined the cognitive mechanisms supporting these
beliefs across Eastern (Japanese) and Western (United States
and Western Europe) cultures, revealing the generality versus
specificity of cognitive mechanisms proposed in the literature
surrounding ESBs. A number of key findings emerged from the
current set of studies. First, consistent with past work (Shiah
et al., 2010), Easterners, such as Japanese participants, showed
stronger paranormal beliefs (Studies 1–3) and some evidence
of stronger pseudoscientific beliefs (Studies 1 and 2) compared
to Western participants8. Conversely, receptivity to pseudo-
profound bullshit did not differ across cultures.
Individual differences in holistic cognition partly explained
differences in ESBs and bullshit receptivity. Specifically,
individual differences in the holistic understanding of
causality (i.e., multiple and complex causality) played a key
role in explaining differences in ESBs and bullshit receptivity.
However, contrary to expectations, participants’ attitudes toward
contradictions failed to explain ESBs and bullshit receptivity,
suggesting that the stronger endorsement of ESBs in our Japanese
sample was not due to a cultural difference in receptiveness to
contradiction. Furthermore, individual differences in thinking
style also explained differences in ESBs and bullshit receptivity,
with a propensity to engage in analytic thinking negatively
associated with the endorsement of ESBs and pseudo-profound
bullshit statements as profound. Interestingly, the association
between Type-2 analytic thinking (as measured by the CRT
and rationality subscale of the REI) and ESBs differed across
cultures. That is, for Westerners, individuals high in analytic
thinking were less likely to endorse ESBs and pseudo-profound
bullshit whereas no association was found between analytic
thinking and either ESBs or bullshit receptivity in our Japanese
sample. Thus, the present findings suggest that a propensity to
engage in analytic thinking may protect against ESBs in Western
individuals, but may not do so across cultures (specifically
Eastern cultures).
Cultural psychologists have proposed a framework to
understand cultural differences in cognitive styles between
8However, it is worth noting that participants’ endorsement of ESBs in the
present research may have been affected by factors other than the believability of
statements. For example, past work has found evidence of a midpoint preference
among Japanese participants (Heine and Lehman, 1999). Thus, factors not related
to participants’ endorsement of ESBs may have nevertheless contributed to these
differences.
Western and Eastern individuals, distinguishing between analytic
and holistic cognition (Nisbett et al., 2001;Nisbett and
Miyamoto, 2005;Ishii, 2013). Related to this distinction, we
hypothesized that individuals demonstrating a high tolerance
for contradictions would be more likely to endorse ESBs, as
we suspected that those accepting contradictions would tend to
perceive pro-ESB statements as plausible even when they did
not specifically endorse the ESB themselves. However, our results
suggest that it is not a tendency to accept contradictions that
positively relates to the endorsement of ESBs, but rather the
consideration that causality is complex and multifaceted and
that events may have multiple causes which need not necessarily
contradict. Holistic thinkers, therefore, may be less impacted by
conflicts caused by scientifically unsound claims and be more
open to alternative explanations of events.
The culture (i.e., holistic cognition) hypothesis may be
challenged by the fact that, with the exception of Study
3, our results failed to show significant sample by holistic
cognition interactions. Nevertheless, it is possible that the
present findings arose, at least in part, due to the mixture
of falsifiable and unfalsifiable statements within ESB items.
That is, the endorsement of an empirically falsifiable belief
(e.g., a pseudoscientific belief) may be more strongly associated
with an analytical mode of thought compared to unfalsifiable
beliefs, on account of falsifiability being connected to the type
of scientific, deductive, linear thinking that is characteristic
of analytic thinking. Consistent with this claim, the observed
negative association between CRT and ESBs was generally
stronger when evaluating participants’ endorsement of falsifiable
pseudoscientific beliefs compared to less readily falsifiable
paranormal ones.
Additionally, within three of our four studies, we observed
a sample by CRT interaction on ESBs and bullshit receptivity.
Therefore, individuals’ CRT performance (an assessment of
their tendency to engage in Type-2 analytic thinking) related
differently to endorsement of ESBs and receptiveness to pseudo-
profound bullshit across cultures. Consistent with past work
(Pennycook et al., 2012, 2015a), CRT performance was negatively
associated with ESBs and bullshit receptivity within our Western
sample, suggesting that more deliberative modes of thinking
may protect against these seemingly irrational beliefs. However,
a different pattern of results was observed in our Japanese
samples, for which CRT was unrelated to the endorsement of
ESBs and pseudo-profound bullshit. Therefore, the present study
suggests that the often observed negative association between
Type-2 analytic thinking and various ESBs may be exclusive to
Western individuals.
Of course, it remains an open question as to why Type-
2 analytic thinking relates differently to the endorsement of
ESBs across cultures. One possibility is that measures of analytic
thinking (such as the CRT) in part measure a tendency to oppose
cultural norms. In support of this claim is the finding that
cognitive reflection is positively associated with religious disbelief
within highly religious countries but shares no association with
religious disbelief in less religious countries (Gervais et al., 2018).
If true, we may expect the association between cognitive reflection
and ESBs to differ between cultures with different cultural norms
Frontiers in Psychology | www.frontiersin.org 13 February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 14
Majima et al. Culture and Epistemically Suspect Beliefs
surrounding ESBs. Notably, in the present study, the effects of
CRT on pseudoscientific beliefs were observed to be roughly the
same across cultures, as opposed to paranormal beliefs which
differed across cultures. One possible explanation for the different
results observed for pseudoscientific and paranormal beliefs is
that both Western and Eastern individuals may hold a similar
norm against endorsing readily falsifiable pseudoscientific claims,
while a norm against endorsing less readily falsifiable paranormal
beliefs may be stronger among Western individuals. Therefore,
in the present context, the weak association between cognitive
reflection and ESBs (particularly paranormal beliefs) in our
Japanese sample may arise due to the fact that these beliefs may
not necessarily violate prevailing cultural norms, while the same
may not be true of a Western sample.
SUMMARY
The present research investigates cultural differences in ESBs
and bullshit receptivity. Critically, we demonstrate how the
importance of Type-2 analytic thinking for suppressing ESBs
(often observed in Western samples) fails to generalize across
cultures (i.e., Japanese samples). Of course, the current work is
not intended to be definitive; therefore, future studies should
explore how factors such as analytic thinking predict other
important real-world beliefs (e.g., endorsement of conspiracy
theories and “fake news”) across cultures. Furthermore, the
underlying mechanisms explaining what cultural factors elicit
cultural differences regarding the endorsement of ESBs remains
an open question. Nevertheless, we provide initial evidence
demonstrating that not only does the endorsement of ESBs differ
across cultures, but so too might the cognitive mechanisms
supporting these beliefs, calling into question the generality of
various cognitive mechanisms often thought to support ESBs.
DATA AVAILABILITY STATEMENT
Data and analysis scripts of all studies are available at https:
//osf.io/cbfwj/.
ETHICS STATEMENT
The studies involving human participants were reviewed and
approved by the Research Ethics Committee at Hokusei Gakuen
University and the Office of Research Ethics, at the University
of Waterloo. All participants provided their written informed
consent to participate in this study.
AUTHOR CONTRIBUTIONS
YM designed the project, conducted empirical studies and
statistical analyses, and drafted the manuscript. AW and MT co-
designed and conducted Study 4 and revised the manuscript. JF
supervised the project and critically reviewed the manuscript. All
authors discussed the results and commented on the manuscript.
FUNDING
This research was supported by grants from JSPS KAKENHI
Grant (15K04033 and 19K03194) and The Natural Sciences and
Engineering Research Council of Canada.
ACKNOWLEDGMENTS
The authors thank Aki Nishihara for her help in translating
materials of Study 1. The authors also thank Crimson Interactive
Pvt. Ltd. (Ulatus) – www.ulatus.jp for their assistance in
translating materials of Study 4.
SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found
online at: https://www.frontiersin.org/articles/10.3389/fpsyg.
2022.745580/full#supplementary-material
REFERENCES
Aarnio, K., and Lindeman, M. (2005). Paranormal beliefs,
education, and thinking styles. Pers. Indiv. Diff. 39, 1227–1236.
doi: 10.1016/j.paid.2005.04.009
Bahçekapili, H. G., and Yilmaz, O. (2017). The relation between different types
of religiosity and analytic cognitive style. Pers. Indiv. Diff. 117, 267–272. doi:
10.1016/j.paid.2017.06.013
Blackmore, S., and Tro´
scianko, T. (1985). Belief in the
paranormal: probability judgements, illusory control, and
the ‘chance baseline shift’. Br. J. Psychol. 76, 459–468.
doi: 10.1111/j.2044-8295.1985.tb01969.x
Bruder, M., Haffke, P., Neave, N., Nouripanah, N., and Imhoff, R. (2013).
Measuring individual differences in generic beliefs in conspiracy theories across
cultures: conspiracy mentality questionnaire. Front. Psychol. 4:225. doi: 10.
3389/fpsyg.2013.00225
Buchtel, E. E., and Norenzayan, A. (2009). “Thinking across cultures: implications
for dual processes,” in In Two Minds: Dual Processes and Beyond, eds
J. S. B. T. Evans and K. Frankish (Oxford: Oxford University Press),
217–238.
Choi, I., Koo, M., and Choi, J. A. (2007). Individual differences in analytic
versus holistic thinking. Pers. Soc. Psychol. Bull. 33, 691–705. doi: 10.1177/
0146167206298568
Cohen, P., Cohen, J., Aiken, L. S., and West, S. G. (1999). The problem of units and
the circumstance for POMP. Multivar. Behav. Res. 34, 315–346. doi: 10.1207/
S15327906MBR3403_2
Dekker, S., Lee, N. C., Howard-Jones, P., Jolles, J., Kuan-Chung, C.,
and Jordan, A. (2012). Neuromyths in education: prevalence and
predictors of misconceptions among teachers. Front. Psychol. 3:429.
doi: 10.3389/fpsyg.2012.00429
Epstein, S., Pacini, R., Denes-Raj, V., and Heier, H. (1996).
Individual differences in intuitive-experiential and analytical-
rational thinking styles. J. Pers. Soc. Psychol. 71, 390–405.
doi: 10.1037/0022-3514.71.2.390
Evans, J. S. B. T. (2008). Dual-processing accounts of
reasoning, judgment, and social cognition. Annu. Rev.
Frontiers in Psychology | www.frontiersin.org 14 February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 15
Majima et al. Culture and Epistemically Suspect Beliefs
Psychol. 59, 255–278. doi: 10.1146/annurev.psych.59.103006.
093629
Evans, J. S. B. T., and Stanovich, K. E. (2013). Dual-process theories of higher
cognition: advancing the debate. Perspect. Psychol. Sci. 8, 223–241. doi: 10.1177/
1745691612460685
Fagerlin, A., Zikmund-Fisher, B. J., Ubel, P. A., Jankovic, A., Derry, H. A., and
Smith, D. M. (2007). Measuring numeracy without a math test: development
of the subjective numeracy scale. Med. Dec. Mak. 27, 672–680. doi: 10.1177/
0272989x07304449
Frederick, S. (2005). Cognitive reflection and decision making. J. Econ. Perspect. 19,
25–42. doi: 10.1257/089533005775196732
Gervais, W. M., and Norenzayan, A. (2012). Analytic
thinking promotes religious disbelief. Science 336, 493–496.
doi: 10.1126/science.1215647
Gervais, W. M., van Elk, M., Xygalatas, D., McKay, R. T., Aveyard, M., Buchtel,
E. E., et al. (2018). Analytic atheism: a cross-culturally weak and fickle
phenomenon? Judg. Dec. Mak. 13, 268–274.
Gray, T., and Mill, D. (1990). Critical abilities, graduate education (Biology vs.
English), and belief in unsubstantiated phenomena. Can. J. Behav. Sci. 22,
162–172. doi: 10.1037/h0078899
Heine, S. J., and Lehman, D. R. (1999). Culture, self-discrepancies, and
self-satisfaction. Pers. Soc. Psychol. Bull. 25, 915–925. doi: 10.1177/
01461672992511001
Henrich, J., Heine, S. J., and Norenzayan, A. (2010). The weirdest people in the
world? Behav. Brain Sci. 33, 61–83. doi: 10.1017/S0140525X0999152X
Ishii, K. (2013). Culture and the mode of thought: a review. Asian J. Soc. Psychol.
16, 123–132. doi: 10.1111/ajsp.12011
Karasawa, K., and Tsukimoto, T. (2010). The effect of information processing
style on beliefs toward paranormal phenomena. J. Hum. Environ. Stud. 8, 1–5.
doi: 10.4189/shes.8.1
Lindeman, M. (2011). Biases in intuitive reasoning and belief in complementary
and alternative medicine. Psychol. Health 26, 371–382. doi: 10.1080/
08870440903440707
Lindeman, M., and Aarnio, K. (2006). Paranormal beliefs: their
dimensionality and correlates. Eur. J. Pers. 20, 585–602.
doi: 10.1002/per.608
Lobato, E., Mendoza, J., Sims, V., and Chin, M. (2014). Examining the relationship
between conspiracy theories, paranormal beliefs, and pseudoscience acceptance
among a university population. Appl. Cogn. Psychol. 28, 617–625. doi: 10.1002/
acp.3042
Maddux, W. W., and Yuki, M. (2006). The “Ripple Effect”: cultural differences in
perceptions of the consequences of events. Pers. Soc. Psychol. Bull. 32, 669–683.
doi: 10.1177/0146167205283840
Majima, Y. (2015). Belief in pseudoscience, cognitive style
and science literacy. Appl. Cogn. Psychol. 29, 552–559.
doi: 10.1002/acp.3136
Markovits, H., and Nantel, G. (1989). The belief-bias effect in the production
and evaluation of logical conclusions. Mem. Cogn. 17, 11–17. doi: 10.3758/
BF03199552
Miyamoto, Y., Uchida, Y., and Ellsworth, P. C. (2010). Culture and mixed
emotions: co-occurrence of positive and negative emotions in Japan and the
United States. Emotion 10, 404–415. doi: 10.1037/a0018430
Naito, M., Suzuki, K., and Sakamoto, A. (2004). Development of rational and
intuitive information-processing style inventory. Jpn. J. Pers. 13, 67–78. doi:
10.2132/personality.13.67
Nakajima, S., Sato, T., and Watanabe, Y. (1992). Tobacyk no Chou shizen genshou
shinpou shakudo ni tuite. [On the Tobacyk’s Paranormal Belief Scale]. J. Japan
Skept. 1, 73–79.
Nakajima, S., Sato, T., and Watanabe, Y. (1993). Chou shizen genshou shinpou
shakudo no sakusei [the development of belief in paranormal phenomena
scale]. J. Japan Skept. 2, 69–80.
Nisbett, R. E., Choi, I., Peng, K., and Norenzayan, A. (2001). Culture and systems
of thought: holistic versus analytic cognition. Psychol. Rev. 108, 291–310. doi:
10.1037/0033-295X.108.2.291
Nisbett, R. E., and Miyamoto, Y. (2005). The influence of culture: holistic versus
analytic perception. Trends Cogn. Sci. 9, 467–473. doi: 10.1016/j.tics.2005.08.
004
Norenzayan, A., Smith, E. E., Kim, B. J., and Nisbett, R. E. (2002). Cultural
preferences for formal versus intuitive reasoning. Cogn. Sci. 26, 653–684. doi:
10.1207/s15516709cog2605_4
Pacini, R., and Epstein, S. (1999). The relation of rational and experiential
information processing styles to personality, basic beliefs, and the ratio-bias
phenomenon. J. Pers. Soc. Psychol. 76, 972–987. doi: 10.1037/0022-3514.76.6.
972
Pennycook, G., Cheyne, J. A., Barr, N., Koehler, D. J., and Fugelsang, J. A. (2014).
Cognitive style and religiosity: the role of conflict detection. Mem. Cogn. 42,
1–10. doi: 10.3758/s13421-013- 0340-7
Pennycook, G., Cheyne, J. A., Barr, N., Koehler, D. J., and Fugelsang, J. A. (2015a).
On the reception and detection of pseudo-profound bullshit. Judg. Dec. Mak.
10, 549–563.
Pennycook, G., Cheyne, J. A., Seli, P., Koehler, D. J., and Fugelsang, J. A. (2012).
Analytic cognitive style predicts religious and paranormal belief. Cognition 123,
335–346. doi: 10.1016/j.cognition.2012.03.003
Pennycook, G., Fugelsang, J. A., and Koehler, D. J. (2015b). Everyday consequences
of analytic thinking. Curr. Dir. Psychol. Sci. 24, 425–432. doi: 10.1177/
0963721415604610
Pennycook, G., and Rand, D. G. (2020). Who falls for fake news? The roles of
bullshit receptivity, overclaiming, familiarity, and analytic thinking. J. Pers. 88,
185–200.
Risen, J. L. (2016). Believing what we do not believe: acquiescence to superstitious
beliefs and other powerful intuitions. Psychol. Rev. 123, 182–207. doi: 10.1037/
rev0000017
Roberts, M. J., and Seager, P. B. (1999). Predicting belief in paranormal
phenomena: a comparison of conditional and probabilistic reasoning. Appl.
Cogn. Psychol. 13, 443–450. doi: 10.1002/(SICI)1099-0720(199910)13:5<443::
AID-ACP592<3.0.CO;2-K
Sanchez, C., Sundermeier, B., Gray, K., and Calin-Jageman, R. J. (2017). Direct
replication of Gervais & Norenzayan (2012): No evidence that analytic thinking
decreases religious belief. PLoS One 12:e0172636. doi: 10.1371/journal.pone.
0172636
Shiah, Y.-J., Wu, M.-H., Tam, W.-C. C., and Chang, F. (2010).
Paranormal beliefs and religiosity: chinese version of the
revised paranormal belief scale. Psychol. Rep. 107, 367–382.
doi: 10.2466/08.09.17.Pr0.107.5.367-382
Spencer-Rodgers, J., Peng, K., Wang, L., and Hou, Y.
(2004). Dialectical self-esteem and East-West differences in
psychological well-being. Pers. Soc. Psychol. Bull. 30, 1416–1432.
doi: 10.1177/0146167204264243
Spencer-Rodgers, J., Williams, M. J., and Peng, K. (2010). Cultural differences in
expectations of change and tolerance for contradiction: a decade of empirical
research. Pers. Soc. Psychol. Rev. 14, 296–312. doi: 10.1177/1088868310362982
Tobacyk, J. J. (2004). A revised paranormal belief scale. Int. J. Transpers. Stud. 23,
94–98.
Toplak, M. E., West, R. F., and Stanovich, K. E. (2011). The cognitive reflection test
as a predictor of performance on heuristics-and-biases tasks. Mem. Cogn. 39,
1275–1289. doi: 10.3758/s13421-011- 0104-1
Tosyali, F., and Aktas, B. (2021). Does training analytical thinking decrease
superstitious beliefs? Relationship between analytical thinking, intrinsic
religiosity, and superstitious beliefs. Pers. Indiv. Differ. 183:111122. doi: 10.
1016/j.paid.2021.111122
Turpin, M. H., Walker, A. C., Kara-Yakoubian, M., Gabert, N. N., Fugelsang, J. A.,
and Stolz, J. A. (2019). Bullshit makes the art grow profounder. Judg. Dec. Mak.
14, 658–670.
van Prooijen, J.-W., Douglas, K. M., and De Inocencio, C. (2018). Connecting
the dots: Illusory pattern perception predicts belief in conspiracies and the
supernatural. Eur. J. Soc. Psychol. 48, 320–335. doi: 10.1002/ejsp.2331
Walker, A. C., Turpin, M. H., Stolz, J. A., Fugelsang, J. A., and Koehler, D. J. (2019).
Finding meaning in the clouds: illusory pattern perception predicts receptivity
to pseudo-profound bullshit. Judg. Dec. Mak. 14, 109–119.
Whitson, J. A., and Galinsky, A. D. (2008). Lacking control increases illusory
pattern perception. Science 322, 115–117. doi: 10.1126/science.1159845
Yong, J. C., Hartanto, A., and Tan, J. J. X. (2021). Subjective social status and
inflammation: the role of culture and anger control. Health Psychol. 40, 62–70.
doi: 10.1037/hea0001029
Frontiers in Psychology | www.frontiersin.org 15 February 2022 | Volume 13 | Article 745580

fpsyg-13-745580 February 3, 2022 Time: 14:52 # 16
Majima et al. Culture and Epistemically Suspect Beliefs
Yonker, J. E., Edman, L. R. O., Cresswell, J., and Barrett, J. L.
(2016). Primed analytic thought and religiosity: the importance
of individual characteristics. Psychol. Relig. Spirit. 8, 298–308.
doi: 10.1037/rel0000095
Zhang, B., Galbraith, N., Yama, H., Wang, L., and Manktelow, K. I.
(2015). Dialectical thinking: a cross-cultural study of Japanese,
Chinese, and British students. J. Cogn. Psychol. 27, 771–779.
doi: 10.1080/20445911.2015.1025792
Conflict of Interest: The authors declare that the research was conducted in the
absence of any commercial or financial relationships that could be construed as a
potential conflict of interest.
Publisher’s Note: All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated organizations, or those of
the publisher, the editors and the reviewers. Any product that may be evaluated in
this article, or claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.
Copyright © 2022 Majima, Walker, Turpin and Fugelsang. This is an open-access
article distributed under the terms of the Creative Commons Attribution License
(CC BY). The use, distribution or reproduction in other forums is permitted, provided
the original author(s) and the copyright owner(s) are credited and that the original
publication in this journal is cited, in accordance with accepted academicpractice. No
use, distribution or reproduction is permitted which does not comply with theseterms.
Frontiers in Psychology | www.frontiersin.org 16 February 2022 | Volume 13 | Article 745580