报告
To Believe Is Not to Think: A Cross-Cultural Finding
Neil Van Leeuwen1, Kara Weisman2*, and Tanya Marie Luhrmann3
1Department of Philosophy and Neuroscience Institute, Georgia State University
2心理学系, 加州大学, Riverside
3Department of Anthropology, 斯坦福大学
*Kara Weisman was at Stanford University (Psychology and Anthropology Departments) at the time of the writing of this article.
开放访问
杂志
关键词: 信仰, 思维, credence, cognitive attitudes, epistemic verbs, religious psychology, 理论
of mind
抽象的
Are religious beliefs psychologically different from matter-of-fact beliefs? Many scholars say
不: that religious people, in a matter-of-fact way, simply think their deities exist. Others say
是的: that religious beliefs are more compartmentalized, less certain, and less responsive to
证据. Little research to date has explored whether lay people themselves recognize such a
difference. We addressed this question in a series of sentence completion tasks, conducted in
five settings that differed both in religious traditions and in language: 美国, 加纳,
Thailand, 中国, and Vanuatu. Participants everywhere routinely used different verbs to
describe religious versus matter-of-fact beliefs, and they did so even when the ascribed belief
contents were held constant and only the surrounding context varied. These findings support
the view that people from diverse cultures and language communities recognize a difference
in attitude type between religious belief and everyday matter-of-fact belief.
我
D
哦
w
n
哦
A
d
e
d
F
r
哦
米
H
t
t
p
:
/
/
d
我
r
e
C
t
.
米
我
t
.
/
e
d
你
哦
p
米
我
/
我
A
r
t
我
C
e
–
p
d
F
/
d
哦
我
/
我
/
.
/
1
0
1
1
6
2
哦
p
米
_
A
_
0
0
0
4
4
1
9
6
3
6
0
0
哦
p
米
_
A
_
0
0
0
4
4
p
d
.
/
我
F
乙
y
G
你
e
s
t
t
哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3
介绍: VARIETIES OF BELIEF
Many people have beliefs about gods, 祖先, souls, and other supernatural phenomena.
的确, religious beliefs have shaped human history for millennia. But people also have beliefs
about whether it will rain, where their children are, or what time of year it is; such matter-of-
fact beliefs are central to everyday life.
Do people’s religious beliefs typically involve the same cognitive attitude as their matter-of-
fact beliefs? The phrase “cognitive attitudes” refers to the various ways humans process and
relate to ideas about the world (Shah & Velleman, 2005; Van Leeuwen, 2014). A person might
think that it’s about to rain, wonder whether it’s about to rain, assume, 怀疑, imagine, pre-
tend, 希望, or worry that it’s about to rain. At stake here is whether religious ideas—for all
their importance—are usually processed in the same way as the matter-of-fact belief that it
is about to rain.
Many scholars say the cognitive attitudes are the same (Boudry & Coyne, 2016A, 2016乙; S.
Harris et al., 2009; 征收, 2017; Talmont-Kaminski, 2016). On this view, religious people just
think gods and spirits exist, just like anyone simply thinks tables and chairs exist. Others hold
that religious beliefs are different: more compartmentalized (Astuti & 哈里斯, 2008; 磷. L. 哈里斯
引文: Van Leeuwen, N。, Weisman,
K., & Luhrmann, 时间. 中号. (2021). To Believe
Is Not to Think: A Cross-Cultural
Finding. 开放的心态: Discoveries in
认知科学, 5, 91–99. https://土井
.org/10.1162/opmi_a_00044
DOI:
https://doi.org/10.1162/opmi_a_00044
补充材料:
https://doi.org/10.1162/opmi_a_00044;
https://osf.io/qy3js/; https://github.com
/ kgweisman/think_believe
已收到: 4 十一月 2020
公认: 27 六月 2021
利益争夺:
The authors declare no conflict
of interest.
通讯作者:
Neil Van Leeuwen
nvan@gsu.edu
版权: © 2021
麻省理工学院
在知识共享下发布
归因 4.0 国际的
(抄送 4.0) 执照
麻省理工学院出版社
Believe Is Not to Think
Van Leeuwen et al.
& Giménez, 2005; Watson-Jones et al., 2016), more effortful (博耶, 2013; Luhrmann, 2018),
in need of regular voluntary practice (Luhrmann, 2012; Norenzayan, 2013), less certain (Clegg
等人。, 2019; Davoodi et al., 2018; 磷. L. Harris et al., 2006), less responsive to counterevidence
(Liquin et al., 2020; Stanovich & Toplak, 2019; Van Leeuwen, 2017), and experienced as part
of one’s identity (Durkheim, 1912/2008).
This debate concerns not the contents of religious vs. matter-of-fact belief but the attitude, 或者
the way people relate to contents or ideas. Here is a philosophical characterization of this
difference: For any content p (supernatural or naturalistic; scientific or not; observable or
unobservable, ETC。), one can relate to that content in various ways. Consider the following
attitude reports:
1) Karen doubts that it will rain.
2) Sarah hopes that it will rain.
3) Bob doubts that God exists.
4) Arthur hopes that God exists.
The italicized attitude contents reported in 1 和 2 are naturalistic, 尽管 3 和 4 concern the
supernatural. But the attitudes of doubting and hoping can occur with either kind of content.
Now consider the attitudes one might hold toward these contents:
Jane factually believes that Jesus was born in Bethlehem.
5)
6) Fred has the religious belief that Jesus was born in Bethlehem.
Despite having the same contents, one might suggest that the second mental state, 6, is a rever-
ential, identity-constituting attitude that is compartmentalized, less responsive to evidence, 和
less widely used than factual belief to support ordinary inferences. One of us ( Van Leeuwen,
2014), 例如, has argued that this is a much more typical attitude for religious contents.
其他的, 然而, such as Levy (2017), deny that there is such a difference in “belief” type:
Religious beliefs are just factual beliefs that happen to be about supernatural contents (因此
the title of his paper: “Religious Beliefs Are Factual Beliefs”). So the dispute comes down to
whether a meaningfully different cognitive attitude of religious credence exists at all.
在本文中, we present new data that speak to this important debate. Our studies tested
whether lay people themselves recognize a difference between matter-of-fact and religious
信仰. We predicted that people from diverse cultural and religious settings would systemat-
ically choose different words for describing matter-of-fact vs. religious cognitive attitudes,
thereby revealing that they distinguish them.
迄今为止, one series of studies has addressed such a prediction. Heiphetz et al. (2021) 用过的
corpus analyses as well as sentence completion tasks to show that American English speakers are
more likely to use the word “believe” to describe religious beliefs (例如, “Zane believes that Jesus
turned water into wine”), but “think” to describe matter-of-fact beliefs (例如, “Nick thinks that
George Washington was the first U.S. president”). 重要的, this was so even when they held
reported attitude contents constant and varied the situation to make the surrounding context
religious or nonreligious. They thus concluded not only that there are distinct attitudes, 但是也
that lay people are at some level aware of the difference and use available resources to express it.
These earlier studies, 然而, left an important question open: How widespread is this
distinction across cultures? Many observers hold that “religious belief” is only a Western,
Christian, and perhaps modern idea: Other people do not believe—they know (Asad, 1993;
开放的心态: 认知科学的发现
92
我
D
哦
w
n
哦
A
d
e
d
F
r
哦
米
H
t
t
p
:
/
/
d
我
r
e
C
t
.
米
我
t
.
/
e
d
你
哦
p
米
我
/
我
A
r
t
我
C
e
–
p
d
F
/
d
哦
我
/
我
.
/
/
1
0
1
1
6
2
哦
p
米
_
A
_
0
0
0
4
4
1
9
6
3
6
0
0
哦
p
米
_
A
_
0
0
0
4
4
p
d
.
/
我
F
乙
y
G
你
e
s
t
t
哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
史密斯, 1977). In a representative statement, Toren (2007, p. 307) writes: “我们 [anthropologists]
may characterise as belief what our informants know and, in so doing, misrepresent them.”
The implication is that the distinction surfaced in Heiphetz et al. (2021) does not appear in
non-Western, non-Christian contexts. Some anthropologists, 然而, have found that reli-
gious belief is indeed a distinct attitude in various non-Western cultures—for example, 这
Vezo in Madagascar and the Fang in Central Africa (Astuti & 哈里斯, 2008; 博耶, 2013).
Following recent calls to assess psychological theories in a variety of cultural settings
(Henrich et al., 2010), we set out to test empirically whether people differentiate matter-of-fact
and religious belief in culturally diverse fieldsites. The studies we present here were part of a
larger project investigating cultural models of the mind and their relationship to spiritual ex-
periences, which took place in five countries—from west to east: 美国, 加纳,
Thailand, 中国, and Vanuatu—chosen to include a variety of spiritual and religious practices
(for more details on the project, see Luhrmann, 2020; Luhrmann et al., 2021). 参加者
were primarily Christian or religiously unaffiliated in the United States, overwhelmingly
Christian in Ghana and Vanuatu, Buddhist in Thailand, and religiously unaffiliated in China.
For purposes of designing our study stimuli, we used a “minimum definition” of religion that
would be useful cross-culturally: 想法, talk, and practices that concern gods, spirits, or other
supernatural beings (比照. Goody, 1961). Even in religious traditions that seem to have no spirits
(such as Buddhism), participants often have ideas and practices about a wide range of spirits
(总的, 2002). 那就是说, we used this definition provisionally, and we hold open the possibility
that in some cultural contexts people’s religious attitudes can concern other subject matters as
出色地 (see General Discussion).
在美国, studies were conducted in English and featured the words think and
believe, following Heiphetz et al. (2021). For other countries, counterparts to “think” and “be-
lieve” were chosen in consultation with native speakers and anthropologists with local exper-
tise. These counterparts were not expected to be exact semantic matches. 相当, we predicted
that people would use them in parallel ways to track the distinction between religious and
matter-of-fact belief. In Ghana, studies were conducted in Fante (an Akan dialect) and focused
on the words dwen and gye dzi. In Thailand, studies were conducted in Thai and focused on
(chu’(西德:1)A). In China, studies were conducted in Mandarin (in the variety
the words
known as Standard Chinese) and focused on the words 认为 (rènwéi) and 相信 (xia(西德:1)ngxìn). 在
Vanuatu, studies were conducted in Bislama (an English-based creole) and focused on the words
ting and bilif. All study materials were back-translated to ensure accuracy.
(kit) 和
To be clear: Our main interest in these studies was not in the semantics of these pairs of
words per se, but in the way that people use them to report distinct cognitive attitudes. 我们的
hypothesis was that the ability to do so is widespread across cultures; we predicted that people
would think differently about religious and matter-of-fact beliefs and that these different under-
standings would manifest as differential use of words like “think” and “believe” in sentence
completion tasks. We have not in these studies addressed the following questions: (1) 如何
many other ways can the words we focused on be used, aside from the ways we focus on
这里? (2) To what extent do the verb pairs we focus on have resembling semantic profiles
across languages, aside from being close enough to facilitate a test of our prediction? (3)
How did cultural influences shape the semantics and pragmatics of these words over historical
时间? Beyond providing a test for the hypothesis we have fleshed out here, we hope that the
studies that follow provide a launching point for investigating those related questions.
Our three studies, which closely follow Heiphetz et al.’s (2021) experimental studies, each
tested whether lay people use the pairs of words indicated here to communicate a difference
开放的心态: 认知科学的发现
93
我
D
哦
w
n
哦
A
d
e
d
F
r
哦
米
H
t
t
p
:
/
/
d
我
r
e
C
t
.
米
我
t
.
/
e
d
你
哦
p
米
我
/
我
A
r
t
我
C
e
–
p
d
F
/
d
哦
我
/
我
/
.
/
1
0
1
1
6
2
哦
p
米
_
A
_
0
0
0
4
4
1
9
6
3
6
0
0
哦
p
米
_
A
_
0
0
0
4
4
p
d
/
.
我
F
乙
y
G
你
e
s
t
t
哦
n
0
7
S
e
p
e
米
乙
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
between matter-of-fact and religious beliefs. For each study, we predicted that participants
would be more likely to use “believe” (or its hypothesized counterpart) for religious belief than
for matter-of-fact belief. (Our studies were preregistered at https://aspredicted.org/p6iy3.pdf;
see the Supplemental Materials for complete preregistered analyses, as well as further details
on method.)
STUDY 1: FORCED CHOICE (“THINK” VS. “BELIEVE”)
Study 1 provided an initial test of our prediction.
参加者 (N= 344; n = 48–97 per site) were presented with 25 sentences in one of two
counterbalanced orders. Each sentence had the form “[Character] [thinks / believes] that X,”
where X was a propositional phrase (例如, 约翰 [thinks / believes] that Jesus Christ died for
human sins). Participants selected one of the two words to complete the sentence. 在里面
美国, Thailand, 中国, and Vanuatu, participants completed a pen-and-paper survey.
In Ghana, the study was administered orally at a rural site, since Fante is rarely written; at this
地点, a research assistant read each item out loud and recorded verbal responses.
Ten of these sentences were “religious”: The complement phrase included Christian con-
帐篷 (例如, Jesus Christ died for human sins) or Buddhist content (例如, the Buddha found spir-
itual truth while meditating).
The remaining 15 sentences were “matter-of-fact”: The complement included a widely
known fact (例如, Brazil is in South America), a less widely known fact (例如, bronze contains
more copper than tin), or a personal life fact (例如, her dad is cooking noodles for dinner).
A mixed effects logistic regression revealed that, as predicted, participants were generally
more likely to select “believe” (or its counterpart) to complete religious vs. matter-of-fact atti-
tude ascriptions (β = .20, p < .001). This distinction was more pronounced in Thailand and the
United States, and less pronounced in Ghana; it did not differ from the grand mean in China or
Vanuatu (see Table S3 for complete results). Nonetheless, secondary analyses confirmed that
this difference was significant in each fieldsite considered alone (United States: β = .24, p <
.001; Ghana: β = .09, p = .010; Thailand: β = .27, p < .001; China: β = .19, p < .001; Vanuatu:
β = .20, p < .001; see Table S4 and Figure 1).
Figure 1. Study 1 results. Participants in all fieldsites were more likely to select “believe” (or its
counterparts in other languages) to complete religious vs. matter-of-fact sentences.
OPEN MIND: Discoveries in Cognitive Science
94
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
4
4
1
9
6
3
6
0
0
o
p
m
_
a
_
0
0
0
4
4
p
d
.
/
i
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
STUDY 2: FREE RESPONSE
In Study 2, a separate group of participants (N = 388; n = 46–100 per site) completed the same
sentences using a word or phrase of their own free choice. Methods were otherwise identical
to Study 1.
Echoing Study 1, participants were more likely to generate responses containing the word-
stem “believe” (or its counterpart) for religious compared to matter-of-fact attitude ascriptions
(β = .13, p < .001). Indeed, in four of these five sites, “believe” (or its counterpart) appeared in
the plurality of free responses to religious sentences; see Figure 2. The difference between re-
ligious and matter-of-fact sentences was more pronounced in the United States and China, less
pronounced in Ghana and Vanuatu, and did not vary from the grand mean among participants
in Thailand (see Table S11). But again, this difference was significant in each fieldsite consid-
ered alone (United States: β = .20, p < .001; Ghana: β = .06, p < .001; Thailand: β = .14, p <
.001; China: β = .17, p < .001; Vanuatu: β = .09, p < .001; see Table S12).
STUDY 3: VIGNETTE COMPLETION
Study 3 (N = 328; n = 49–80 per site) provided a final, closely controlled test of our primary
claim. Methods were similar to Study 1, except that, rather than completing attitude ascriptions
that were either religious or matter-of-fact in content, there were five pairs of brief vignettes that
set up either a matter-of-fact or religious context, but ended with the same final sentence that
the participant had to complete. For example:
Kerry had bad headaches in the afternoons all last year. Sometimes her friends offered
her aspirin. But Kerry took courses at a medical school. That school teaches that drinking
water is the way to cure a headache and aspirin is not. So Kerry always refused the aspirin
her friends offered. That’s because she [ thinks / believes ] that aspirin is not a cure.
Terry had bad headaches in the afternoons all last year. Sometimes her friends offered
her aspirin. But Terry belonged to the Church of Christ Scientist. That church teaches that
prayer is the way to cure illness and medicine is not. So Terry always refused the aspirin
her friends offered. That’s because she [ thinks / believes ] that aspirin is not a cure.
Study 2 results. Participants in all fieldsites were more likely to generate responses containing the word stem “believe” (or its
Figure 2.
counterparts) to complete religious vs. matter-of-fact sentences. This plot presents 13 word stems that include the six most common responses
in each fieldsite, ordered by prevalence.
OPEN MIND: Discoveries in Cognitive Science
95
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
4
4
1
9
6
3
6
0
0
o
p
m
_
a
_
0
0
0
4
4
p
d
.
/
i
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
Figure 3. Study 3 results. Participants in the United States, Thailand, China, and Vanuatu—but not Ghana—were more likely to select
“believe” (or counterparts) to complete an attitude report when it was embedded in a religious (vs. matter-of-fact) vignette.
This allowed us to test ascriptions of cognitive attitudes while holding the ascribed content
constant. In this case, Kerry and Terry might be ascribed different attitudes to the same content:
that aspirin is not a cure.
Vignettes were presented in one of two counterbalanced orders, with paired vignettes sep-
arated from each other. “Religious” vignettes included diverse religious traditions, and “matter-
of-fact” vignettes were scientific, historical, or commonsensical in nature.
As predicted, participants were generally more likely to select “believe” (or its counterpart)
when the sentence was embedded in a religious vignette as opposed to the matching matter-of-
fact vignette (β = .10, p = .010). This distinction was more pronounced in the United States and
Thailand, less pronounced in Ghana, and did not differ from the grand mean among partici-
pants in China or Vanuatu (see Table S20). The difference was significant in four of the five sites
considered alone (United States: β = .15, p < .001; Thailand: β = .15, p = .009; China: β = .11, p =
.033; Vanuatu: β = .08, p = .049; see Table S21 and Figure 3). The difference was not significant
in our sample from rural Ghana (β = .02, p = .747), but it did appear to be present in a sample of
Ghanaian undergraduates in a small follow-up study (see the Supplemental Materials).
GENERAL DISCUSSION
To step back, across Studies 1 and 2, participants were generally more likely to use the word
“believe” (or its counterparts in other languages) to describe cognitive attitudes that relate to
gods, ancestors, souls, and other supernatural phenomena, than to describe matter-of-fact be-
liefs. In Study 3, which featured cognitive attitudes with neutral contents, participants were
more likely to use “believe” when the surrounding vignette made clear that the attitude in
question was held in a religious context than when the surrounding vignette provided a
matter-of-fact context.
We consider this to be evidence that matter-of-fact beliefs and religious beliefs involve dis-
tinct cognitive attitudes, that awareness of the difference is widespread among human cultural
OPEN MIND: Discoveries in Cognitive Science
96
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
/
.
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
4
4
1
9
6
3
6
0
0
o
p
m
_
a
_
0
0
0
4
4
p
d
/
.
i
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
groups, and that this distinction manifests in the word choices of people speaking a wide variety
of languages.
Across all our studies and sites, there was only one exception to the pattern that held every-
where else: Among Ghanaian participants, preferential use of gye dzi (the Fante counterpart to
“believe”) for religious as opposed to matter-of-fact attitudes was present, but attenuated, in
Studies 1–2, and absent in Study 3. This may simply be an experimental artifact, but it raises
the intriguing possibility that in Ghana—a setting where thought and talk about the supernatural
is quite commonplace (Dulin, 2020; Dzokoto, 2020)—many people hold more matter-of-fact
attitudes about religious ideas (see the Supplemental Materials for further discussion).
Conversely, there may also be cultural settings or spheres of discourse in which people hold
“religious” belief attitudes about factual content; for example, recent years have made it clear
that beliefs about such “here-and-now” matters as disease transmission and election security
can become central parts of people’s identities and can be extremely difficult to counter with
factual evidence.
Our studies ruled out possible alternate explanations for the main pattern we found (dis-
cussed in more detail in the Supplemental Materials).
First, if participants used the word “believe” (or counterparts) merely to indicate that the
protagonist of the sentence or vignette was uncertain or unconfident, then we would likely
have seen greater incidence of “believe” for less well-known content—items designated a
priori as “less-widely-known facts”—than for well-known facts in Studies 1–2 (for a discussion
of the relation between belief confidence and perceived consensus, see Shtulman, 2013,
p. 207). But no such pattern emerged, so mere ascription of lesser confidence is unlikely to be
what participants were indicating with their differential word choice.
Second, if participants used the word “believe” only to indicate that they themselves dis-
agreed with the attributed attitude, then we likely would have found an effect of religion, with
people using “believe” (or counterparts) more for religions other than their own. No such pat-
tern emerged. In fact, particularly in the most devoutly Christian samples—in Ghana and
Vanuatu—people were even more likely to use “believe” in sentences with content from
their own religion.
Third, if “believe” reflected only religious content (rather than a difference in cognitive
attitude), then we wouldn’t have seen the differences that we saw in Study 3, in which prop-
ositional complements were matched exactly and only the broader context varied.
In fact, Study 3 also addressed a range of further content-based explanations of the effects
that surfaced in Studies 1 and 2. One might, for example, argue that the effects in Studies 1 and
2 emerged because people were using “believe” for unobservable subjects and “think” for
observable subjects, or that they were using “believe” for ideas that were matters of opinion
and “think” for objective matters. But for Study 3, these distinctions can’t be the explanation:
with attitude contents held fixed (e.g., that aspirin is not a cure), there could not be a difference
in what the respective attitudes were about.
Nonetheless, these studies raise a range of interesting questions for future research. Might
differential use of verbs like “believe” vs. “think” also indicate differences between moral and
nonmoral cognitive attitudes, between beliefs arrived at reflectively vs. intuitively, or between
more and less speculative cognitive attitudes in science? Finding such patterns would not con-
flict with the results here; rather, they would complement our overall outlook by affirming that
people in general have the cognitive flexibility to understand and express nuanced differences
in cognitive attitude type.
OPEN MIND: Discoveries in Cognitive Science
97
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
.
/
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
4
4
1
9
6
3
6
0
0
o
p
m
_
a
_
0
0
0
4
4
p
d
/
.
i
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
Fourth and finally, if the differentiation between matter-of-fact belief and religious belief
were unique to Western contexts, we would not have seen the striking similarities across cul-
tures that we did.
Our results are most parsimoniously explained by our main hypothesis: Matter-of-fact belief
and religious belief are distinct cognitive attitudes, and people in many different cultures and
language communities are aware of the difference. The cognitive flexibility needed to utilize
and differentiate these attitudes is not specific to Westerners, Christians, scholars, or some other
rarefied group; instead, it appears to be widely shared. Matter-of-fact beliefs are likely used in a
problem-solving way to achieve practical goals, while religious beliefs are used in guiding
symbolic actions expressive of sacred values (Atran & Axelrod, 2008); thus, tracking the dis-
tinction between them may allow people to better understand and predict others’ behaviors.
Indeed, this distinction may be one of the common features of a theory of mind that we in-
creasingly understand to be subtle and sophisticated across social worlds.
ACKNOWLEDGMENTS
Thanks especially to the team leaders at each fieldsite: Joshua Brahinsky and Nikki Ross-
Zehnder ( United States), John Dulin and Vivian Dzokoto (Ghana), Felicity Aulino
(Thailand), Emily Ng (China), and Rachel Smith ( Vanuatu). Thanks to Dan Weiskopf for insight
into the hypotheses tested here, to Larisa Heiphetz for help with study design and preregistra-
tion, and to Jonathan Jong, Kilu von Prince, and Rebecca Tuvel for feedback on earlier drafts.
We also thank the editors of Open Mind and three anonymous referees for feedback. See the
Supplemental Materials for further acknowledgments.
FUNDING INFORMATION
TL, John Templeton Foundation (https://dx.doi.org/10.13039/100000925), Award ID: 55427.
KW, National Science Foundation (https://dx.doi.org/10.13039/100000001), Award ID:
DGE-114747.
AUTHOR CONTRIBUTIONS
NVL: Conceptualization: Lead; Investigation: Equal; Methodology: Equal; Writing – Original
Draft: Lead; Writing – Review & Editing: Equal. KW: Data Curation: Lead; Formal Analysis: Lead;
Validation: Lead; Visualization: Lead; Writing – Original Draft: Supporting; Writing – Review &
Editing: Equal. TL: Conceptualization: Supporting; Investigation: Equal; Methodology: Equal;
Project Administration: Lead; Supervision: Lead; Writing – Original Draft: Supporting; Writing –
Review & Editing: Equal.
REFERENCES
Asad, T. (1993). Genealogies of religion: Discipline and reasons of
power in Christianity and Islam. Johns Hopkins University Press.
Astuti, R., & Harris, P. (2008). Understanding mortality and the life of
the ancestors in rural Madagascar. Cognitive Science, 32, 713–740.
https://doi.org/10.1080/03640210802066907, PubMed:
21635351
Atran, S., & Axelrod, R. (2008). Reframing sacred values.
Negotiation Journal, 24(3), 221–246. https://doi.org/10.1111/j
.1571-9979.2008.00182.x
Boudry, M., & Coyne, J. (2016a). Disbelief in belief: On the cognitive
status of supernatural beliefs. Philosophical Psychology, 29(4),
601–615. https://doi.org/10.1080/09515089.2015.1110852
Boudry, M., & Coyne, J. (2016b). Fakers, fanatics, and false dilemmas:
Reply to Van Leeuwen. Philosophical Psychology, 29(4), 622–627.
https://doi.org/10.1080/09515089.2016.1146244
Boyer, P. (2013). Why “belief” is hard work: Implications of
Tanya Luhrmann’s When God talks back. Hau, Journal of
Ethnographic Theory, 3(3), 349–357. https://doi.org/10.14318
/hau3.3.015
Clegg, J. M., Cui, Y. K., Harris, P. L., & Corriveau, K. H. (2019). God,
germs, and evolution: Belief in unobservable religious and scien-
tific entities in the US and China. Integrative Psychological and
Behavioral Science, 53(1), 93–106. https://doi.org/10.1007
/s12124-019-9471-0, PubMed: 30729421
OPEN MIND: Discoveries in Cognitive Science
98
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
.
/
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
4
4
1
9
6
3
6
0
0
o
p
m
_
a
_
0
0
0
4
4
p
d
.
/
i
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Believe Is Not to Think
Van Leeuwen et al.
Davoodi, T., Jamshidi-Sianaki, M., Abedi, F., Payir, A., Cui, Y. K.,
Harris, P. L., & Corriveau, K. H. (2018). Beliefs about religious
and scientific entities among parents and children in Iran.
Social Psychological and Personality Science, 10(7), 847–855.
https://doi.org/10.1177/1948550618806057
Dulin, J. (2020). Vulnerable minds, bodily thoughts, and sensory
spirits: Local theory of mind and spiritual experience in Ghana.
Journal of the Royal Anthropological Institute, 26(S1), 61–76.
https://doi.org/10.1111/1467-9655.13241
Durkheim, É. (2008). The elementary forms of religious life (Carol
Cosman, Trans.). Oxford University Press. (Original work pub-
lished 1912)
Dzokoto, V. (2020). Adwenhoasem: An Akan theory of mind.
Journal of the Royal Anthropological Institute, 26(S1), 77–94.
https://doi.org/10.1111/1467-9655.13242
Goody, J. (1961). Religion and ritual: The definitional problem. The
British Journal of Sociology, 12(2), 142–164. https://doi.org/10
.2307/586928
Gross, R. M. (2002). Meditation and prayer: A comparative inquiry.
Buddhist-Christian Studies, 22, 77–86. https://doi.org/10.1353
/bcs.2002.0009
Harris, P. L., & Giménez, M. (2005). Children’s acceptance of
conflicting testimony: The case of death. Journal of Cognition
and Culture, 5(1–2), 143–164. https://doi.org/10.1163
/1568537054068606
Harris, P. L., Pasquini, E. S., Duke, S., Asscher, J. J., & Pons, F.
(2006). Germs and angels: The role of testimony in young chil-
dren’s ontology. Developmental Science, 9(1), 76–96. https://doi
.org/10.1111/j.1467-7687.2005.00465.x, PubMed: 16445398
Harris, S., Kaplan, J. T., Curiel, A., Bookheimer, S. Y., Iacoboni, M.,
& Cohen, M. S. (2009). The neural correlates of religious and
non-religious belief. PLoS ONE, 4(10), Article e0007272.
https://doi.org/10.1371/journal.pone.0007272, PubMed:
19794914
Heiphetz, L., Landers, C. L., & Van Leeuwen, N. (2021). Does think
mean the same thing as believe? Linguistic insights into religious
cognition. Psychology of Religion and Spirituality, 13(3), 287–297.
https://doi.org/10.1037/rel0000238
Henrich, J., Heine, S., & Norenzayan, A. (2010). The weirdest peo-
ple in the world? Behavioral and Brain Sciences, 33(2–3), 61–83.
https://doi.org/10.1017/S0140525X0999152X, PubMed:
20550733
Levy, N. (2017). Religious beliefs are factual beliefs: Content does
not correlate with context sensitivity. Cognition, 161, 109–116.
https://doi.org/10.1016/j.cognition.2017.01.012, PubMed:
28161595
Liquin, E. G., Metz, S. E., & Lombrozo, T. (2020). Science demands
explanation, religion tolerates mystery. Cognition, 204, Article
104398. https://doi.org/10.1016/j.cognition.2020.104398,
PubMed: 32711182
Luhrmann, T. (2012). When God talks back. Vintage Books.
Luhrmann, T. (2018). The faith frame: Or, belief is easy, faith is
hard. Contemporary Pragmatism, 15(3), 302–318. https://doi
.org/10.1163/18758185-01503003
Luhrmann, T. M. (Ed.) (2020). Mind and spirit: A comparative the-
ory [Special issue]. Journal of the Royal Anthropological Institute,
26(S1), 1–166. https://doi.org/10.1111/1467-9655.13238
Luhrmann, T. M., Weisman, K., Aulino, F., Brahinsky, J. D., Dulin,
J. C., Dzokoto, V. A., Legare, C. H., Lifshitz, M., Ng, E., Ross-
Zehnder, N., & Smith, R. E. (2021). Sensing the presence of gods
and spirits across cultures and faiths. Proceedings of the National
Academy of Sciences, 118(5), Article e2016649118. https://doi
.org/10.1073/pnas.2016649118, PubMed: 33495328
Norenzayan, A. (2013). Big gods: How religion transformed coop-
eration and conflict. Princeton University Press. https://doi.org/10
.1515/9781400848324
Shah, N., & Velleman, J. D. (2005). Doxastic deliberation.
Philosophical Review, 114(4), 497–534. https://doi.org/10.1215
/00318108-114-4-497
Shtulman, A. (2013). Epistemic similarities between students’ scien-
tific and supernatural beliefs. Journal of Educational Psychology,
105(1), 199–212. https://doi.org/10.1037/a0030282
Smith, W. C. (1977). Believing—An historical perspective.
Oneworld Publications.
Stanovich, K. E., & Toplak, M. E. (2019). The need for intellectual
diversity in psychological science: Our own studies of actively
open-minded thinking as a case study. Cognition, 187, 156–166.
https://doi.org/10.1016/j.cognition.2019.03.006, PubMed:
30877847
Talmont-Kaminski, K. (2016). Commentary: Religious credence is
not factual belief. Frontiers in Psychology, 7, Article 1544.
https://doi.org/10.3389/fpsyg.2016.01544, PubMed: 27790166
Toren, C. (2007). How do we know what is true? The case of mana
in Fiji. In R. Astuti, J. P. Parry, & C. Stafford (Eds.), Questions of
anthropology (pp. 307–335). Berg. https://doi.org/10.4324
/9781003086482-12
Van Leeuwen, N. (2014). Religious credence is not factual belief.
Cognition, 133(3), 698–715. https://doi.org/10.1016/j.cognition
.2014.08.015, PubMed: 25268465
Van Leeuwen, N. (2017). Do religious “beliefs” respond to evi-
dence? Philosophical Explorations, 20(sup1), 52–72. https://doi
.org/10.1080/13869795.2017.1287294
Watson-Jones, R. E., Busch, J. T. A., Legare, C. H., & Harris, P.
(2016). Does the body survive death? Cultural variation in beliefs
about life everlasting. Cognitive Science, 41(S3), 455–476.
https://doi.org/10.1111/cogs.12430, PubMed: 27859566
OPEN MIND: Discoveries in Cognitive Science
99
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
4
4
1
9
6
3
6
0
0
o
p
m
_
a
_
0
0
0
4
4
p
d
/
.
i
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
