Religious People Are Trusted Because They Are Viewed as Slow Life-History Strategists

Abstract

Religious people are more trusted than nonreligious people. Although most theorists attribute these perceptions to the beliefs of religious targets, religious individuals also differ in behavioral ways that might cue trust. We examined whether perceivers might trust religious targets more because they heuristically associate religion with slow life-history strategies. In three experiments, we found that religious targets are viewed as slow life-history strategists and that these findings are not the result of a universally positive halo effect; that the effect of target religion on trust is significantly mediated by the target’s life-history traits (i.e., perceived reproductive strategy); and that when perceivers have direct information about a target’s reproductive strategy, their ratings of trust are driven primarily by his or her reproductive strategy, rather than religion. These effects operate over and above targets’ belief in moralizing gods and offer a novel theoretical perspective on religion and trust.

Full text

https://doi.org/10.1177/0956797617753606
Psychological Science
2018, Vol. 29(6) 947 –960
© The Author(s) 2018
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0956797617753606
www.psychologicalscience.org/PS
Research Article
Research has consistently demonstrated that religious
targets are viewed as more trustworthy than nonreligious
targets (e.g., Tan & Vogel, 2008), and, similarly, atheists
tend to be distrusted (Gervais, Shariff, & Norenzayan,
2011). Gervais and colleagues (2017) found that, across
13 nations, the majority of people—even atheists—tend
to associate immoral behavior with atheism.
The prevailing view on religion, irreligion, and trust
posits that intuitions about religion and trust are the result
of cultural group selection, with shared beliefs in punish-
ing deities allowing large-scale cooperation to evolve
(Norenzayan et al., 2016; Purzycki et al., 2016; Roes &
Raymond, 2003). In this view, religious behavior should
facilitate trust primarily toward coreligionists, par-
ticularly when it is diagnostic of belief in moralizing
deities. This framework has garnered significant empiri-
cal support (e.g., Shariff, Willard, Andersen, & Norenzayan,
2016).
Here, we propose an additional reason why religious
behavior might cultivate trust. Specifically, religious indi-
viduals tend to differ from nonreligious individuals in
more than merely belief—they also tend to behave in
ways consistent with a slow life-history (LH) strategy (i.e.,
they tend to be sexually restricted, invested in family,
nonimpulsive, and nonaggressive; Baumard & Chevallier,
2015). These traits are associated with cooperativeness
and prosociality. One possibility, then, is that perceivers
use religious affiliation or behavior as a cue to infer these
traits. Perception of these traits may facilitate trust above
and beyond shared belief in supernatural punishment
and may be particularly important in explaining why
religious behavior can also facilitate trust in out-group
perceivers (Hall, Cohen, Meyer, Varley, & Brewer, 2015;
McCullough, Swartwout, Carter, Shaver, & Sosis, 2016).
An LH Framework
Evolutionary biology’s LH theory has been used to
account for a wide range of behavior, both in humans
753606PSSXXX10.1177/0956797617753606Moon et al.Religion, Life History, and Trust
research-article2018
Corresponding Author:
Jordan W. Moon, Arizona State University, Department of Psychology,
950 S. McAllister Ave., Tempe, AZ 85287
E-mail: jordan.w.moon@asu.edu
Religious People Are Trusted Because
They Are Viewed as Slow Life-History
Strategists
Jordan W. Moon , Jaimie Arona Krems, and Adam B. Cohen
Department of Psychology, Arizona State University
Abstract
Religious people are more trusted than nonreligious people. Although most theorists attribute these perceptions to
the beliefs of religious targets, religious individuals also differ in behavioral ways that might cue trust. We examined
whether perceivers might trust religious targets more because they heuristically associate religion with slow life-history
strategies. In three experiments, we found that religious targets are viewed as slow life-history strategists and that these
findings are not the result of a universally positive halo effect; that the effect of target religion on trust is significantly
mediated by the target’s life-history traits (i.e., perceived reproductive strategy); and that when perceivers have direct
information about a target’s reproductive strategy, their ratings of trust are driven primarily by his or her reproductive
strategy, rather than religion. These effects operate over and above targets’ belief in moralizing gods and offer a novel
theoretical perspective on religion and trust.
Keywords
evolutionary psychology, religious beliefs, religion, life-history theory, trust, open data, open materials
Received 9/16/17; Revision accepted 12/18/17

948 Moon et al.
and other animals (Stearns, 1992). This theory considers
the trade-offs individuals make to navigate the chal-
lenges unique to their environment. In a harsh or unpre-
dictable environment, where the future is less certain,
a fast LH strategy makes more sense—fast strategists
tend to mate early and frequently, discount the future
more steeply, display more aggression, and engage in
riskier behavior that may have more immediate payoffs
(Frankenhuis, Panchanathan, & Nettle, 2016). In con-
trast, individuals who come from stable, predictable
ecologies tend to adopt a slow LH strategy, entailing a
later sexual debut, fewer sexual partners, greater paren-
tal investment, lower levels of aggression and risk-
taking, and greater investment in education. These traits
and behaviors tend to cluster reliably together on a
fast-slow continuum (Ellis, Figueredo, Brumbach, &
Schlomer, 2009) and represent functional adaptations to
specific ecologies (Frankenhuis et al., 2016).
LH strategies have important implications for social-
ity, as extended prosociality and social trust are risky
for actors in unpredictable environments (Petersen &
Aarøe, 2015; Wilson, O’Brien, & Sesma, 2009). Not sur-
prisingly, then, fast strategies and associated traits (e.g.,
temporal discounting, risk taking, unrestricted sexual-
ity) are highly correlated with several noncooperative
or antisocial behaviors (Curry, Price, & Price, 2008; Ellis
et al., 2012; Jonason, Li, & Webster, 2009).
Social perceivers are keenly aware of the implica-
tions of LH strategy on their interactions with others,
as they intuitively link LH cues (e.g., ecology) with
suites of corresponding LH traits (Williams, Sng, &
Neuberg, 2016). Because trust is highly related to coop-
eration (Simpson, 2007), one might expect individuals
perceived as fast LH strategists to be considered poor
cooperators and, thus, less trustworthy than those
viewed as slow strategists.
Religious Individuals as Slow
LH Strategists
Recent work has linked religiosity to slow LH strategies
(for a review, see Baumard & Chevallier, 2015). Religious
individuals tend to display many of the traits evoked by
predictable ecologies, including lower levels of temporal
discounting, greater self-control, and sexual restricted-
ness (Carter, McCullough, Kim-Spoon, Corrales, & Blake,
2012; McCullough & Willoughby, 2009; Schmitt & Fuller,
2015; Weeden, Cohen, & Kenrick, 2008; Weeden &
Kurzban, 2013). Religious individuals are much more
likely to follow a committed reproductive strategy,1
pursuing high-commitment, monogamous relation-
ships, fewer sexual partners, and high parental invest-
ment. Some theorists even view religion and moral
intuitions as a consequence—rather than a cause—of
the reproductive or LH strategies people adopt (Gladden,
Welch, Figueredo, & Jacobs, 2009; Weeden & Kurzban,
2013).
The Present Work
One way perceivers might ascertain the likely disposi-
tion and cooperative value of others is through LH infor-
mation, and religion may serve as a valuable heuristic
about the LH tactics and, thus, trustworthiness of others.
In three experiments, we examined whether religious
targets are viewed as slow LH strategists and, if so,
whether this effect might partly explain why religious
individuals are generally deemed more trustworthy. In
Experiment 1, we demonstrated that religious (com-
pared with nonreligious) targets are viewed as slower
in several specific dimensions of LH strategy (i.e., repro-
ductive strategy, impulsivity, aggression, education, and
ecology) and were also deemed more trustworthy. In
Experiment 2, we manipulated targets’ reproductive
strategies (a significant mediator in Experiment 1) in
addition to their religion. We found that direct informa-
tion about LH strategy (i.e., the target’s “dating prefer-
ence”) tended to override the effects of religious
information (cf. Williams et al., 2016). In Experiment 3,
we attempted to generalize these results, finding that
perceived reproductive strategy statistically mediated
trust toward both Christian and Muslim targets in a pro-
fessional domain. These results suggest that perceivers
utilize religion as an LH cue and trust religious people
more because of the (slow) LH strategy they are assumed
to adopt.
Experiment 1
Experiment 1 had two aims: to examine whether per-
ceptions of religious targets might accurately be catego-
rized as slow LH perceptions and, if so, whether these
perceptions might influence trust.
Method
Participants. We conducted a pilot study to estimate
the effect of religious information on perceived LH traits.
Given this effect size, we determined we would need 196
participants to detect our effect. As our resources allowed,
our final sample exceeded this target. A total of 336 (195
female) workers on Amazon’s Mechanical Turk (MTurk)
located in the United States received $0.25 each to com-
plete a survey on “social perceptions.” Participants who
had taken part in the pilot study were not able to partici-
pate. Participants’ ages ranged from 19 to 71, with a median
of 33 (M = 36.25, SD = 12.73). Forty-five additional par-
ticipants were excluded from the original sample of 381

Religion, Life History, and Trust 949
because they failed at least one of two attention checks—
either answering a question that they were instructed to
skip or failing a multiple-choice question that simply
instructed participants to click “2.” (In the former case,
participants could not unclick the question buttons to
refrain from answering questions, so those who realized
this error and later acknowledged it in their comments to
researchers at the end of the experiment were treated as
passing this check.)
Procedure and measures. Participants completed a
three-item religiosity scale (1 = not at all, 7 = deeply or
extremely; e.g., “How strongly do you believe in God?”
α = .95) based on the work of Cohen, Malka, Rozin, and
Cherfas (2006). They also responded to two additional
measures included for a separate research question: the
attitude facet subscale of the revised Sociosexual Orienta-
tion Inventory (SOI-R; e.g., “Sex without love is okay,”
α = .76) on a 1 (strongly disagree) to 9 (strongly agree)
scale (Penke & Asendorpf, 2008), and five items assessing
self-perceived mate value (e.g., “Members of the opposite
sex are attracted to me,” α = .88) on a 1 (strongly dis-
agree) to 7 (strongly agree) scale (Landolt, Lalumière, &
Quinsey, 1995).
Adapting the methodology of Hall et al. (2015), we
randomly assigned participants to view one of four
fictitious dating profiles, varying in target sex (male or
female) and target religion (“Devoted Christian” vs.
“Nonreligious”). After viewing the target dating profile,
participants answered questions about their perceptions
of the target on several dimensions (in random order),
using a 7-point scale (1 = extremely unlikely, 7 =
extremely likely). Finally, participants entered demo-
graphic information, including age, gender, sexual ori-
entation, and religious affiliation.
We computed scores for five separate (perceived)
LH variables: reproductive strategy (seven items; e.g.,
“Faithful romantic partner,” α = .90), non-aggression
(three items; e.g., “Physically aggressive,” reverse coded,
α = .86), nonimpulsivity (two items; e.g., “Impulsive,”
reverse coded, r = .32, p < .001), education (two items;
e.g., “Invested in his/her education,” r = .78, p < .001),
and hopeful ecology (single item, “comes from a rough
neighborhood,” reverse coded). We also included one
item in which we expected religious targets to be
viewed as “fast”—how likely they are to want several
children (Rowthorn, 2011). Finally, we assessed per-
ceived mate value (four items; e.g., “Gets a lot of
attention from women/men,” α = .90) and trust (single
item).
Results
Are religious targets viewed as slow strategists?. We
compared perceptions of religious and nonreligious targets
in each LH domain (i.e., reproductive strategy, aggression,
impulsivity, education, and ecology) as well as mate value
and trust (see Fig. 1). We were interested in target sex and
participant sex only to the extent that they might moder-
ate the hypothesized main effect of religion on perceived
3
4
5
6
Reproductive Strategy Nonaggression Nonimpulsivity Education Hopeful Ecology Mate Value Trust
Participant Rating
Perceived Life-History Traits, Mate Value, and Trust
Religious Target Nonreligious Target
Fig. 1. Results from Experiment 1: mean life-history inferences, estimated mate value, and trust as a function of target religious claims.
Error bars represent ±1 SE.

950 Moon et al.
LH traits. Because we failed to find evidence that either
target or participant sex moderated the effect, we report
the results of independent-samples t tests comparing reli-
gious and nonreligious targets (see Table 1). For analyses
including these variables, see Supplemental Analyses in
the Supplemental Material.
As predicted, religious (vs. nonreligious) targets were
judged more likely to follow a committed reproductive
strategy, less impulsive, less aggressive, more educated,
and originating from more hopeful ecologies (i.e., less
likely to come from a “rough neighborhood”). Consis-
tent with past research, religious targets were also
trusted significantly more than nonreligious targets.
Importantly, religious targets were not favored univer-
sally, as they were judged similarly in mate value. We
consider this to imply that effects of target religion on
perceived LH traits are not due to an overall halo effect.
Do LH intuitions mediate effects on trust?. We per-
formed a parallel multiple mediation analysis (Hayes,
2013) in which we tested whether the effect portraying a
target person as religious had on participant ratings of
trust was significantly mediated by perceived LH traits. We
used 5,000 bootstrapped iterations to compute a bias-cor-
rected 95% confidence interval (CI) for the indirect effects.
As shown in Figure 2, the effect of target religion
on trust was significantly mediated by perceived repro-
ductive strategy, b = 0.36, SE = 0.09, 95% CI = [0.19,
0.55], as well as perceived education, b = 0.09, SE =
0.05, 95% CI = [0.03, 0.21]. Other indirect effects were
not significant, nor was the direct effect, b = 0.07,
SE = 0.11, p > .250. This analysis suggests that religious
individuals are viewed as trustworthy in part because
they are slow LH strategists, and perceived reproduc-
tive strategy seems to be a particularly strong determi-
nant of trust.
Does participant religiosity moderate the effect of
target religion on trust?. Although religious targets
were trusted more than nonreligious targets, one possi-
bility is that this effect was driven solely by religious par-
ticipants. Thus, we tested whether participants’ self-
reported religiosity moderated the effect of target religion
on trust. We regressed target religion (dummy coded),
participant religiosity (centered), and their interaction on
trust. While religiosity had no main effect (p > .250), there
was a main effect of target religion, t(332) = 10.82, p <
.001, b = 1.11, 95% CI = [0.91, 1.32], qualified by a signifi-
cant interaction between target religion and religiosity,
t(332) = 2.92, p = .004, b = 0.17, 95% CI = [0.05, 0.28].
Simple-slopes analyses suggested that the effect was stron-
gest for participants high in religiosity, t(332) = 5.91, p <
.001, b = 0.96, 95% CI = [0.64, 1.28], but remained margin-
ally significant for those low in religiosity, t(332) = 1.78,
p = .076, b = 0.29, 95% CI = [−0.03, 0.61]. That is, even
participants low in religiosity (1.38 on a 1–7 scale) tended
to trust the religious target more than the target who
ostensibly shares their beliefs. This finding is consistent
with recent research suggesting that nonreligious indi-
viduals associate atheism with immorality, albeit to a
lesser extent than more religious individuals (Gervais et
al., 2017).
We additionally used Model 59 of PROCESS (which
allows the moderator to interact with all three paths in
the model; Hayes, 2013) to test whether the indirect
effect of target religion on trust (via perceived repro-
ductive strategy) held at varying levels (±1 SD) of par-
ticipant religiosity. We were specifically interested in
whether this effect held for participants low in religios-
ity. The indirect effect was stronger among more reli-
gious participants, but remained significant at all levels
of participant religiosity: low, b = 0.46, SE = 0.11, 95%
CI = [0.28, 0.71]; average, b = 0.63, SE = 0.08, 95% CI =
Table 1. Effects of Target Religion on Perceived Life-History Traits, Mate Value, and Trust in Experiment 1
Religious target
(n = 166)
Nonreligious
target (n = 170)
t(334) p
95% CI for
the difference
between
means
Cohen’s dMeasure M SD M SD
Reproductive strategy 5.45 1.01 4.33 0.92 10.66 < .001 [0.91, 1.33] 1.16
Nonimpulsivity 4.94 1.06 4.25 1.05 5.96 < .001 [0.46, 0.91] 0.65
Nonaggression 5.60 1.14 5.16 1.19 3.49 < .001 [0.19, 0.69] 0.38
Education 5.39 1.00 5.03 1.20 2.98 .003 [0.12, 0.60] 0.33
Hopeful ecology 5.72 1.21 5.34 1.32 2.76 .006 [0.11, 0.65] 0.30
Mate value 5.37 1.07 5.35 1.14 0.15 .885 [–0.22, 0.25] 0.02
Trust 5.31 1.04 4.68 1.10 5.42 < .001 [0.40, 0.86] 0.59
Note: Each row represents the results of an independent-samples t test comparing perceptions of religious and nonreligious targets.
CI = confidence interval.

Religion, Life History, and Trust 951
[0.48, 0.82]; and high, b = 0.78, SE = 0.13, 95%
CI = [0.54, 1.05]. The direct effect was not significant at
any level of religiosity (p > .246).
Experiment 2
Experiment 1 demonstrated that religious targets are
viewed as slow LH strategists and are more trusted, and
a mediation analysis suggested that religious people
are more trusted because of their perceived slow LH
strategy (especially their reproductive strategy). We rea-
soned that, if perceivers view religious targets as trust-
worthy primarily because they also view them as
committed reproductive strategists, providing direct
information about their committed reproductive strate-
gies may “override” the effects of religious claims (for
a similar design, see Williams et al., 2016). That is, we
hypothesized that perceivers would base their ratings
primarily on direct LH information (i.e., a target’s repro-
ductive strategy) rather than a cue (religion) that imper-
fectly predicts LH information. To test this hypothesis,
we utilized a concurrent double randomization design,
independently manipulating both targets’ religious
claims as well as their reproductive strategies. This
design allowed us to test the causal effect of the pro-
posed mediator (perceived reproductive strategy) on
trust (Pirlott & MacKinnon, 2016). Additionally, we mea-
sured perceptions that the target believed in divine
punishment (“big gods”), to see if effects of LH strategy
on trust would operate over and above the potential
effect of the target’s perceived belief in big gods.
Method
Participants. We sought to obtain at least 100 partici-
pants per cell for a 2 × 2 design, allowing for potential
exclusions based on attention checks. Thus, we recruited
445 participants (203 female, 2 not reported) via TurkPrime
(Litman, Robinson, & Abberbock, 2017) to complete a sur-
vey about “impressions of others” in exchange for $0.50.
Ages ranged from 19 to 72, with a median age of 33 (M =
35.12, SD = 10.37). Participants were restricted to the
United States, and we restricted availability to those who
had not participated in Experiment 1 or our pilot study.
We also excluded an additional 30 participants who
failed at least one of two attention checks instructing
them to select “6” to demonstrate that they were paying
attention; these participants were excluded from all anal-
yses, including the descriptive results above.
Procedure and measures. The procedure was similar
to that used in Experiment 1. Participants first responded
to an eight-item measure of religiosity (based on Cohen
et al., 2006) using a 7-point scale (1 = not at all, 7 = deeply
or extremely; e.g., “How religious are you?” α = .97), the
SOI-R attitude facet (α = .85), and a measure of belief in
Reproductive Strategy
b = 0.36, SE = 0.09
95% CI = [0.19, 0.55]
Target Religion Trust
Nonaggression
b = 0.04, SE = 0.03
95% CI = [−0.01, 0.11]
Nonimpulsivity
b = 0.06, SE = 0.05
95% CI = [−0.02, 0.16]
Education
b = 0.09, SE = 0.05
95% CI = [0.03, 0.21 ]
Hopeful Ecology
b = 0.01, SE = 0.02
95% CI = [−0.03, 0.06]
b = 0.69, p < .001
b = 0.44, p < .001
b = 0.38, p = .006
b = 0.36, p = .003
b = 1.12, p < .001
b = 0.09, p = .112
b = 0.09, p = .068
b = 0.02, p = .591
b = 0.26, p < .001
b = 0.32, p < .001
Fig. 2. Parallel multiple mediation model depicting the effect of target religion on trust as mediated by perceived life-history traits in Experi-
ment 1. The variable for target religious claims was dummy coded (0 = nonreligious, 1 = religious). CI = confidence interval.

952 Moon et al.
“big gods” on a seven-point scale (1 = strongly disagree,
7 = strongly agree; e.g., “I believe God punishes people
for immoral behavior,” α = .96) based on the “big god”
beliefs posited by Norenzayan et al. (2016).
Next, participants randomly viewed a social media
style profile varying in sex (male or female), religious
claims (“Devout Christian” or “Nonreligious”), and “dat-
ing preferences” (“I don’t see myself settling down any
time soon, I enjoy playing the field and meeting a lot
of new people” or “My goal is to find that special some-
one, settle down, and start a family”). Distractor infor-
mation was identical for all conditions.
After viewing the profile, participants indicated how
likely they thought it was that several traits or behaviors
applied to the target, using a Likert-type scale (1 =
extremely unlikely, 7 = extremely likely). We assessed
the following LH perceptions: reproductive strategy
(same scale as Experiment 1, α = .94); nonimpulsivity
(seven-item scale; for example “Acts impulsively,”
reverse coded, α = .93), nonopportunistic behavior
(three-item measure; e.g., “Physically aggressive,”
reverse coded, α = .92), hopeful ecology (three-item
scale; e.g., “Came from a rough neighborhood,” reverse
coded, α = .82), and education (two-item measure; e.g.,
“Invested in his/her education,” r = .82, p < .001). We
also assessed how much targets were perceived to
believe in big gods (using the same beliefs as the par-
ticipant measure, but directed at the target; α = .98)
and trust (six-item measure adapted from Hall et al.,
2015; e.g., “Is trustworthy,” α = .93), and four distractor
items unrelated to LH strategy (i.e., athletic, stylish,
annoying, and reserved).
Results
Collapsing across participant and target sex,2 analyses con-
sisted of a series of 2 (target religion) × 2 (target reproduc-
tive strategy) analyses of variance (ANOVAs)—one for each
outcome measured (see Table 2 and Fig. 3). Measures of
perceived reproductive strategy and belief in moralizing
gods suggest that the manipulations had significant effects:
committed strategists (vs. uncommitted strategists) were
in fact judged more likely to be committed strategists, F(1,
441) = 592.66, p < .001, ηp2 = .573, and so were religious
(vs. nonreligious) targets, F(1, 441) = 22.13, p < .001,
ηp2 = .048. Religious (vs. nonreligious) targets were viewed
as significantly more likely to believe in big gods, F(1, 439) =
589.21, p < .001, ηp2 = .573, and so were committed (vs.
uncommitted) strategists, F(1, 439) = 25.89, p < .001, ηp2 =
.056.
There was a main effect of target religion on per-
ceived impulsivity: Religious targets were judged to be
somewhat less impulsive than nonreligious targets. How-
ever, target reproductive strategy had a considerably
stronger effect: Committed strategists were viewed as
significantly less impulsive than uncommitted strategists.
Further, there were significant main effects of target strat-
egy (but not target religion): Committed (vs. uncommit-
ted) strategists were viewed as “slower” in several
ways—less likely to exhibit opportunistic behavior, more
likely to come from an abundant or predictable ecology,
and more likely to be educated and invested in educa-
tion. Finally, religious targets were viewed as marginally
more trustworthy than nonreligious targets (ηp2 = .008),
but target reproductive strategy had a markedly stronger
effect, with committed strategists being rated as signifi-
cantly more trustworthy (ηp2 = .197).
Do belief in big gods or LH traits mediate the effect
of committed reproductive strategies on trust?. One
possibility is that, although ratings of trust seem to track
reproductive strategy rather than religion, perceivers
assume that committed strategists are more likely to
believe in God or gods. Gervais and colleagues (2017)
found, for instance, that a priest who molested children
was rated likely to be an atheist. Indeed, our analysis did
Table 2. Effects of Target Religion and Target Reproductive Strategy on Perceived Life-History
Traits and Trust in Experiment 2
Target religion
Target reproductive
strategy
Outcome df F ηp2p F ηp2p
Nonimpulsivity 1, 441 16.37 .036 < .001 323.72 .423 < .001
Nonopportunistic behavior 1, 441 2.63 .006 .106 24.31 .052 < .001
Hopeful ecology 1, 441 3.29 .007 .070 7.04 .016 .008
Education 1, 441 0.06 .000 .814 62.02 .123 < .001
Trust 1, 440 3.69 .008 .055 108.15 .197 < .001
Note: Each row represents the results of a 2 (target religion) × 2 (target reproductive strategy) analysis of
variance with the outcome as a dependent variable. The tests reported represent the main effect of each factor,
respectively. Degrees of freedom apply to both main effects. The Target Religion × Target Reproductive Strategy
interaction was not significant for any of the outcomes measured (p > .102).

Religion, Life History, and Trust 953
indicate that committed strategists were judged some-
what more likely to believe in big gods. To test whether
this might explain the effect on trust, we conducted a
mediation analysis to examine whether the main effect of
target reproductive strategy on trust could be explained
partially by their perceived greater belief in big gods.
This analysis used the same approach to mediation as
Experiment 1.
With perceived belief in big gods as the only media-
tor (i.e., reproductive strategy → big god beliefs →
trust), the indirect effect was significant, b = 0.05, SE =
0.02, 95% CI = [0.01, 0.11], suggesting significant media-
tion. The direct effect of reproductive strategy was still
significant, b = 0.96, SE = 0.10, 95% CI = [0.77, 1.16].
To examine whether other LH traits (e.g., impulsivity,
opportunistic behavior) might mediate the effect, we
conducted a multiple mediation analysis to estimate the
effect of reproductive strategy on trust with the follow-
ing mediators: perceived big god beliefs, nonimpulsiv-
ity, nonopportunistic behavior, education, and hopeful
ecology (see Fig. 4). In this model, perceived big god
beliefs did not significantly mediate the effect, b =
0.004, SE = 0.09, 95% CI = [−0.01, 0.03]. However, three
of the four LH variables significantly mediated the effect
of reproductive strategy on trust: nonimpulsivity, b =
0.59, SE = 0.09, 95% CI = [0.43, 0.79]; nonopportunistic
behavior, b = 0.09, SE = 0.03, 95% CI = [0.04, 0.17]; and
education, b = 0.28, SE = 0.05, 95% CI = [0.19, 0.39],
but not hopeful ecology, b = −0.008, SE = 0.01, 95%
CI = [−0.03, 0.01]. The direct effect was not significant
in this model, b = 0.05, SE = 0.09, 95% CI = [−0.13, 0.23].
Does participant religiosity moderate the observed
effect?. We again tested whether participant religiosity
might qualify the observed effects. We regressed target
religion (contrast coded: −0.5 = nonreligious, 0.5 = reli-
gious), target strategy (contrast coded: −0.5 = uncommit-
ted, 0.5 = committed), participant religiosity (centered),
and their interactions on our trust measure. In this model,
there was a significant effect of target religion, t(436) =
2.45, p = .015, b = 0.25, 95% CI = [0.05, 0.44], and a main
effect of target strategy, t(436) = 10.32, p < .001, b = 1.03,
95% CI = [0.83, 1.23]; religious (vs. nonreligious) targets
were trusted more, and committed (vs. uncommitted) tar-
gets were trusted more. There was one significant inter-
action (all other ps > .250) between target strategy and
participant religiosity, t(436) = 3.73, p < .001, b = 0.20,
95% CI = [0.09, 0.30]; more religious participants tended
to rate the committed (vs. uncommitted) strategists as
especially trustworthy. Probing the interaction at low (−1
SD) and high (+1 SD) values of participant religiosity
revealed that participants low in religiosity tended to
trust the committed targets more than the uncommitted
targets, t(436) = 4.65, p < .001, b = 0.66, 95% CI = [0.38,
0.93], and that participants high in religiosity also trusted
the committed targets more than the uncommitted targets
and did so to a greater extent, t(436) = 9.86, p < .001, b =
1.40, 95% CI = [1.13, 1.68]. This finding is consistent with
3
4
5
6
7
Nonimpulsivity Nonopportunistic
Behavior
Education Hopeful Ecology Trust
Participant Rating
Perceived Life-History Traits and Trust
Nonreligious/Uncommitted Religious/Uncommitted Nonreligious/Committed Religious/Committed
Fig. 3. Results from Experiment 2: mean life-history inferences and trust as a function of the target’s religious claims and reproduc-
tive strategy. Error bars represent ±1 SE.

954 Moon et al.
our claim that reproductive strategy largely overrides the
heuristic value of religion.
Experiment 3
Experiments 1 and 2 operationalized the target person’s
religion as Christianity; however, other researchers
(Gervais et al., 2011; Hall et al., 2015) have demon-
strated that both Christian and Muslim targets are
trusted more than nonreligious targets. In Experiment
3, we tested whether perceived reproductive strategies
mediate trust toward both Christian and Muslim targets.
Further, whereas Experiment 1 used dating profiles and
Experiment 2 used social media profiles, we sought to
generalize our results to a professional context.
Method
Participants. Given the smaller effect sizes found in
Experiment 1, we estimated that we would need at least
301 participants for adequate (.80) power. In total, 392
(210 female) participants received $0.50 to complete a
survey on person perception via TurkPrime, again
excluding past participants. Participant ages ranged from
18 to 77, with a median of 32 (M = 34.89, SD = 11.11). We
excluded 20 additional participants from the original
sample of 412 because they failed one or more attention
checks of the same kind used in Experiment 2.
Procedure and measures. Participants completed the
same measures as in Experiment 2: religiosity (α = .96),
SOI-R attitude facet (α = .83), and big god beliefs (α =
.96). They then viewed a randomly assigned profile from
a “business- and employment-oriented social networking
website.” Each profile was a young male listed as an
accountant; under the “What to know about me” section,
he self-identified either as a devoted Christian who
attends church regularly, a devoted Muslim who attends
mosque regularly, or a nonreligious person who does not
follow any particular religion. Other distractor informa-
tion was included and held constant across profiles. To
avoid priming mating, we neither included any informa-
tion about nor explicitly manipulated target reproductive
strategy.
After viewing the profile, participants responded to
the same series of perceived LH measures used in
Experiment 2, again on a 1 (extremely unlikely) to 7
(extremely likely) scale: reproductive strategy (α = .91),
nonimpulsivity (α = .86), nonopportunistic behavior
(α = .94), education (r = .78, p < .001), and hopeful
ecology (α = .76). Participants responded to five face-
valid “accountancy trust” items (e.g., “You would trust
him with your bank account information,” α = .89). Next,
to measure general trust, they responded to the six-item
trust measure used in Experiment 2 (Hall et al., 2015),
which we combined with Cottrell, Neuberg, and Li’s
(2007) established trust measure (e.g., “Honest,” α = .95).
Nonopportunistic
Behavior
b = 0.09, SE = 0.03
95% CI = [0.04, 0.17]
Reproductive Strategy Trust
Big Gods
b = 0.004, SE = 0.03
95% CI = [−0.01, 0.11]
Nonimpulsivity
b = 0.59, SE = 0.09
95% CI = [0.43, 0.79]
Education
b = 0.28, SE = 0.05
95% CI = [0.19, 0.39]
Hopeful Ecology
b = −0.01, SE = 0.01
95% CI = [−0.04, 0.01]
b = 1.74, p < .001
b = 0.55, p = .007
b = 0.28, p = .011
b = 0.84, p < .001
b = 0.55, p < .001
b = 0.34, p < .001
b = 0.01, p = .640
b = 0.34, p < .001
b = 0.17, p < .001
b = –0.03, p = .415
Fig. 4. Parallel multiple mediation model depicting the effect of target reproductive strategy on trust as mediated by perceived life-history
traits and perceived big god beliefs in Experiment 2. The variable for target reproductive strategy was dummy coded (0 = uncommitted, 1 =
committed). CI = confidence interval.

Religion, Life History, and Trust 955
We used the same measure of the target’s perceived belief
in big gods (α = .98) and distractor items as Experiment 2.
Results
As shown in Table 3, target religion had a significant
main effect on general trust, p < .001. Pairwise com-
parisons revealed that both the Christian (p < .001) and
the Muslim (p = .043) targets were judged more trust-
worthy than the nonreligious target. Although the pat-
tern of means was similar in the exploratory accountancy
trust scale (see Fig. 5), the main effect of target religion
was not significant (p > .250), and there were no sig-
nificant pairwise differences (p > .148).
The patterns for other LH variables were similar to
Experiment 1, with some slight exceptions. First, the
Muslim and nonreligious targets were rated similarly
on nonopportunistic behavior and ecology (possibly
because of the inferred racial dimension of being Mus-
lim). Second, target religion had no main effect on
perceived education (p > .250), perhaps because the
target identified himself as an accountant, leaving less
ambiguity. Most importantly, the Christian and Muslim
targets were both rated as more committed reproductive
strategists than the nonreligious target (p < .001).
Do LH perceptions mediate the effect on trust?. Fol-
lowing Hayes and Preacher (2014), we used 5,000 boot-
strapped iterations to compute a bias-corrected 95% CI
for the indirect effects, using two dummy-coded variables
to represent the effect for the Christian condition (nonre-
ligious = 0, christian = 1, Muslim = 0) and the Muslim
condition (nonreligious = 0, Christian = 0, Muslim = 1).
Consistent with Experiment 1, perceived reproduc-
tive strategy arose as a significant mediator of general
trust for both Christian and Muslim targets (see Fig. 6).
Nonimpulsivity was also a significant mediator for both
targets, and nonopportunistic behavior was a significant
mediator for the Christian target only. The direct effect
was not significant for the Christian target, b = −0.10,
SE = 0.09, 95% CI = [−0.27, 0.08], or for the Muslim
target, b = −0.15, SE = 0.09, 95% CI = [−0.32, 0.02].
Using the accountancy trust scale yielded similar
results; reproductive strategy was a significant mediator
for both the Christian, b = 0.34, SE = 0.10, 95% CI =
[0.15, 0.56], and the Muslim target, b = 0.27, SE = 0.08,
95% CI = [0.12, 0.45]. Nonimpulsivity also mediated the
effect for the Christian, b = 0.17, SE = 0.06, 95% CI =
[0.07, 0.31], and the Muslim target, b = 0.11, SE = 0.05,
95% CI = [0.04, 0.23]. Other indirect effects (non-
opportunistic behavior, education, and ecology) were not
significant. The direct effects were slightly, but signifi-
cantly, negative for both the Christian, b = −0.38, SE =
0.15, 95% CI = [−0.68, −0.08], and the Muslim target, b =
0.30, SE = 0.15, 95% CI = [−0.59, −0.02].
Table 3. Effects of Target Religion on Perceived Life-History Traits and Trust in Experiment 3
Nonreligious
target (n = 131)
Christian target
(n = 130)
Muslim target
(n = 131)
Outcome M SD M SD M SD F(2, 389) pη2
Reproductive strategy 4.33a0.85 5.58c0.97 5.31b1.11 58.83 < .001 .230
Nonimpulsivity 4.59a0.99 5.23b0.90 5.02b0.99 15.01 < .001 .072
Nonopportunistic 5.34a1.31 5.82b1.23 5.35a1.36 5.79 .003 .029
Education 5.53a0.98 5.70a1.17 5.72a1.17 1.24 .291 .006
Hopeful ecology 4.81ab 1.04 5.04b1.10 4.61a0.99 5.53 .004 .028
General trust 4.91a0.93 5.44c1.03 5.16b1.04 8.94 < .001 .044
Accountancy trust 4.66a1.32 4.90a1.30 4.78a1.39 1.04 .353 .005
Note: Each row represents the main effect of a one-way analysis of variance comparing perceptions of nonreligious, Christian, and
Muslim targets. Means with different subscripts are significantly different from each other (p < .05).
3
4
5
6
7
General Trust Accountancy Trust
Participant Rating
Trust Measure
Nonreligious Christian Muslim
Fig. 5. Results from Experiment 3: mean trust rating as a function
of target religion. Error bars represent ±1 SE.

956
Reproductive Strategy
bC = 0.35, SE = 0.08
95% CI = [0.22, 0.52]
bM = 0.28, SE = 0.06
95% CI = [0.17, 0.42]
Target Religion General Trust
Nonimpulsivity
bC = 0.13, SE = 0.04
95% CI = [0.06, 0.23]
bM = 0.09, SE = 0.04
95% CI = [0.03, 0.18]
Nonopportunistic
bC = 0.10, SE = 0.04
95% CI = 0.03, 0.19]
bM = 0.002, SE = 0.03
95% CI = [−0.07, 0.07]
Education
bC = 0.03, SE = 0.03
95% CI = [−0.01, 0.10]
bM = 0.04, SE = 0.03
95% CI = [−0.01, 0.10]
Hopeful Ecology
bC = 0.01, SE = 0.01
95% CI = [−0.01, 0.05]
bM = −0.01, SE = 0.01
95% CI = [−0.04, 0.01]
bC = 0.64, p < .001
bM = 0.43, p < .001
bC = 0.48, p = .003
bM = 0.01, p = .960
bC = 1.25, p < .001
bM = 0.98, p < .001
bC = 0.18, p = .197
bM = 0.19, p = .155
bC = 0.22, p = .084
bM = −0.21, p = .111
b = 0.20, p < .001
b = 0.20, p < .001
b = 0.28, p < .001
b = 0.19, p < .001
b = 0.03, p = .278
Fig. 6. Parallel multiple mediation model depicting the effect of target religion on general trust as mediated by perceived life-history traits in Experiment 3. Where
there are two coefficients, bC shows results for the Christian condition, and bM shows results for the Muslim target, relative to the nonreligious condition. CI = confi-
dence interval.

Religion, Life History, and Trust 957
Does participant religiosity moderate the effect?. We
again tested whether participant religiosity moderated
the effect of target religion on trust. Using the general
trust scale and the same dummy-coded variables, the
Christian target was trusted significantly more than the
nonreligious target, t(386) = 4.15, p < .001, b = 0.51, 95%
CI = [0.27, 0.75], and the Muslim target was trusted mar-
ginally more, t(386) = 1.89, p = .059, b = 0.23, 95% CI =
[−0.01, 0.47]. Participant religiosity significantly moder-
ated the effect toward the Christian target, t(386) = 3.45,
p = .001, b = 0.22, 95% CI = [0.09, 0.34], but not toward
the Muslim target (p > .250). Simple-slopes analyses
revealed that participants high in religiosity were more
trusting toward the Christian target, t(386) = 5.53, p <
.001, b = 0.93, 95% CI = [0.60, 1.26], but the effect was not
significant for participants low in religiosity (p > .250).
Using the accountancy trust scale, neither the Chris-
tian (p = .190) nor the Muslim (p > .250) target were
rated more trustworthy than the nonreligious target.
However, religiosity again moderated the effect for the
Christian target, t(386) = 3.79, p < .001, b = 0.32, 95%
CI = [0.15, 0.49]. Simple-slopes analyses revealed that,
among participants high in religiosity, the Christian tar-
get was rated as a more trustworthy accountant than
the nonreligious target, t(386) = 3.72, p < .001, b = 0.83,
95% CI = [0.39, 1.28]. However, among participants low
in religiosity, the Christian target was judged to be a
marginally less trustworthy accountant than the nonre-
ligious target, t(386) = −1.72, p = .086, b = −0.41, 95%
CI = [−0.87, 0.06].
As in Experiment 1, we again used Model 59 of
PROCESS (Hayes, 2013) to test whether the indirect
effect of target religion on general trust (via perceived
reproductive strategy) held across levels (±1 SD) of
participant religiosity, and specifically whether this
effect held even for those participants low in religiosity.
We computed two separate analyses comparing the
nonreligious target to the Christian and to the Muslim,
respectively. For the Christian target, the mediation
effect was significant at the low, b = 0.76, SE = 0.13,
95% CI = [0.52, 1.05]; average, b = 0.81, SE = 0.10, 95%
CI = [0.63, 1.02]; and high, b = 0.77, SE = 0.17, 95%
CI = [0.47, 1.16], levels of religiosity. The direct effect
was significantly negative at the low, b = −0.68, SE =
0.15, 95% CI = [−0.98, −0.38]; and average, b = −0.23,
SE = 0.11, 95% CI = [−0.46, −0.01] levels of religiosity,
but not at high religiosity (b = 0.22, p = .223).
For the Muslim target, the indirect effect was also
significant at the low, b = 0.54, SE = 0.12, 95% CI =
[0.32, 0.79]; average, b = 0.61, SE = 0.08, 95% CI = [0.45,
0.80]; and high, b = 0.66, SE = 0.14, 95% CI = [0.42,
0.99], levels of religiosity. The direct effect was again
significantly negative at the low, b = −0.37, SE = 0.15,
95% CI = [−0.67, −0.08]; average, b = −0.37, SE = 0.11,
95% CI = [−0.58, −0.16]; and high, b = −0.36, SE = 0.16,
95% CI = [−0.68, −0.05], levels of religiosity.
General Discussion
We investigated whether religious targets are viewed
as slow LH strategists and, if so, whether this may
explain effects of religious claims on trust. Experiment
1 demonstrated that religious targets (compared with
nonreligious targets) were rated as “slower” across LH
domains, but were not judged to be more appealing
mates (i.e., it was not a universal “halo” effect). They
were also rated as more trustworthy, and the effect of
target religion on trust was statistically mediated by
perceived reproductive strategy. Experiment 3 showed
that this mediation holds for both Christian and Muslim
targets in a professional domain.
Experiment 2 independently manipulated both the
religion and reproductive strategy of targets to test
whether “direct” LH information might take precedence
over religion in influencing ratings of trust. When par-
ticipants were given targets’ reproductive strategies, the
same religious manipulation used in Experiment 1,
which had a significant effect, became markedly less
important, and target religion had only a marginally
significant effect on trust.
Past research has posited that religious individuals
are trusted in part because they believe they are being
watched by a morally concerned god (Gervais et al.,
2011). Here, we tested a novel explanation for the
robust finding that religious targets tend to be trusted
more than nonreligious targets. Whereas these two
explanations are not mutually exclusive, our data sug-
gest the effects of LH intuitions operate above and
beyond shared beliefs in monitoring gods to cultivate
trust. Consistent with the findings of Hall and col-
leagues (2015), our data suggest that perceivers are
perhaps less concerned with targets’ specific beliefs,
and more interested in the likely behavior they can infer
from religious information. Because religion is not the
only behavior cue (and may not even be the strongest),
the presence of other information may diminish the
effect of religion or, in some cases, override it.
Indeed, our results suggest that simply knowing
someone’s reproductive strategy was sufficient to dimin-
ish drastically the effect of religion. Given the relative
consistency of trust ratings based on religious informa-
tion (both in our data and in past research), it is striking
that any single piece of additional information might
diminish or override this phenomenon. This finding
may present a hopeful picture for anti-atheist bias in
interpersonal contexts. Cottrell and Neuberg (2005)

958 Moon et al.
proposed that prejudices toward different groups are
rooted in the specific threats they are perceived to pose.
Individuals often act in ways to counteract specific prej-
udices—Black men might whistle classical music to self-
present as nonthreatening, and obese individuals might
practice excellent hygiene to avoid being seen as a dis-
ease threat (Neel, Neufeld, & Neuberg, 2013). One simple
way for nonreligious individuals to counteract prejudice,
then, might be to present themselves as family oriented
or invested in monogamous relationships.
Viewing religion as an LH cue may further lead to
nuanced hypotheses about when religious individuals
might not be favored. Given the specific traits that reli-
gion seems to cue, it may be possible to find instances
in which the same religious individuals are viewed posi-
tively (e.g., trustworthy) on one hand, but negatively
(e.g., closed-minded) on the other. Further, whether an
individual views religion favorably or not may depend
on his or her current motives (Cohen & Moon, 2017).
Some researchers have provocatively claimed that
religiosity is essentially a reflection of mating or LH
strategies (Weeden et al., 2008); in this view, religion
is motivated mainly by restricted and monogamous
sexual strategies. In our view, religion is a complex
and nuanced phenomenon that cannot be reduced to
any single motive. Nonetheless, our data suggest that
perceivers may use religion heuristically to infer com-
mitted reproductive strategies and a suite of slow LH
traits. These traits may, in turn, influence perceptions
of trust.
Action Editor
James K. McNulty served as action editor for this article.
Author Contributions
All authors developed the study concept and experimental
design. J. W. Moon collected the data. J. W. Moon analyzed
the data and drafted the manuscript with significant input and
revisions by J. A. Krems and A. B. Cohen.
ORCID iD
Jordan W. Moon https://orcid.org/0000-0001-5102-3585
Declaration of Conflicting Interests
The author(s) declared that there were no conflicts of interest
with respect to the authorship or the publication of this
article.
Funding
This research was supported by the Air Force Office of Sci-
entific Research (Grant No. FA9550-15-1-0008) and a Psi Chi
graduate research grant awarded to J. W. Moon.
Supplemental Material
Additional supporting information can be found at http://
journals.sagepub.com/doi/suppl/10.1177/0956797617753606
Open Practices
All data and materials have been made publicly available via
the Open Science Framework and can be accessed at https://
osf.io/a8sv7/?view_only=cf0e445e30354e22bb53b82242fd94d8.
The experiments were not preregistered. The complete Open
Practices Disclosure for this article can be found at http://jour
nals.sagepub.com/doi/suppl/10.1177/0956797617753606. This
article has received badges for Open Data and Open Materials.
More information about the Open Practices badges can be found
at http://www.psychologicalscience.org/publications/badges.
Notes
1. We use the term “committed reproductive strategy” to denote
the restricted sexual lifestyles associated with religion, and we
view it as a crucial component of a slow LH strategy. However,
slow LH strategists tend to have fewer offspring, whereas reli-
gious individuals tend to have significantly higher fertility rates
than nonreligious individuals (Rowthorn, 2011). We view this
inconsistency as the result of cultural evolution (i.e., religions
that promoted fecundity were more likely to spread; Richerson
& Boyd, 2005; Rowthorn, 2011) and maintain that the psychol-
ogy of these “committed strategists” is fundamentally “slow.”
We view investment in offspring (or quality) as more indicative
of LH strategy than quantity of offspring (Ellis et al., 2009).
For further discussion on fundamental LH trade-offs, see Del
Giudice, Gangestad, and Kaplan (2015).
2. Although we were primarily interested in the effects of tar-
get religion and target reproductive strategy, we manipulated
target sex to be consistent with Experiment 1. Results of the
2 (target sex) × 2 (target religion) × 2 (target reproductive
strategy) for each outcome are reported in the Supplemental
Analyses in the Supplemental Material. The results do not alter
the conclusions.
References
Baumard, N., & Chevallier, C. (2015). The nature and dynam-
ics of world religions: A life-history approach. Proceedings
of the Royal Society B: Biological Sciences, 282, 20151593.
doi:10.1098/rspb.2015.1593
Carter, E. C., McCullough, M. E., Kim-Spoon, J., Corrales,
C., & Blake, A. (2012). Religious people discount the
future less. Evolution and Human Behavior, 33, 224–231.
doi:10.1016/j.evolhumbehav.2011.09.006
Cohen, A. B., Malka, A., Rozin, P., & Cherfas, L. (2006).
Religion and unforgivable offenses. Journal of Personality,
74, 85–117. doi:10.1111/j.1467-6494.2005.00370.x
Cohen, A. B., & Moon, J. W. (2017). Psychology: Atheism
and moral intuitions. Nature Human Behaviour, 1, Article
0157. doi:10.1038/s41562-017-0157

Religion, Life History, and Trust 959
Cottrell, C. A., & Neuberg, S. L. (2005). Different emotional reac-
tions to different groups: A sociofunctional threat-based
approach to “prejudice.” Journal of Personality and Social
Psychology, 88, 770–789. doi:10.1037/0022-3514.88.5.770
Cottrell, C. A., Neuberg, S. L., & Li, N. P. (2007). What do
people desire in others? A sociofunctional perspective
on the importance of different valued characteristics.
Journal of Personality and Social Psychology, 92, 208–231.
doi:10.1037/0022-3514.92.2.208
Curry, O. S., Price, M. E., & Price, J. G. (2008). Patience is
a virtue: Cooperative people have lower discount rates.
Personality and Individual Differences, 44, 778–783.
doi:10.1016/j.paid.2007.09.023
Del Giudice, M., Gangestad, S. W., & Kaplan, H. S. (2015).
Life history theory and evolutionary psychology. In D. M.
Buss (Ed.), The handbook of evolutionary psychology (2nd
ed., pp. 88–114). New York, NY: Wiley.
Ellis, B. J., Del Giudice, M., Dishion, T. J., Gray, P., Hawley, P. H.,
Jacobs, W. J., . . . Wilson, D. S. (2012). The evolutionary
basis of risky adolescent behavior: Implications for sci-
ence, policy, and practice. Developmental Psychology, 48,
598–623. doi:10.1037/a0026220
Ellis, B. J., Figueredo, A. J., Brumbach, B. H., & Schlomer, G. L.
(2009). Fundamental dimensions of environmental risk: The
impact of harsh versus unpredictable environments on the
evolution and development of life history strategies. Human
Nature, 20, 204–268. doi:10.1007/s12110-009-9063-7
Frankenhuis, W. E., Panchanathan, K., & Nettle, D. (2016).
Cognition in harsh and unpredictable environments.
Current Opinion in Psychology, 7, 76–80. doi:10.1016/j
.copsyc.2015.08.011
Gervais, W. M., Shariff, A. F., & Norenzayan, A. (2011). Do
you believe in atheists? Distrust is central to anti-atheist
prejudice. Journal of Personality and Social Psychology,
101, 1189–1206. doi:10.1037/a0025882
Gervais, W. M., Xygalatas, D., McKay, R. T., van Elk, M., Buchtel,
E. E., Aveyard, M., . . . Bulbulia, J. (2017). Global evidence
of extreme intuitive moral prejudice against atheists. Nature
Human Behaviour, 1, Article 0151. doi:10.1038/s41562-
017-0151
Gladden, P. R., Welch, J., Figueredo, A. J., & Jacobs, W. J.
(2009). Moral intuitions and religiosity as spuriously corre-
lated life history traits. Journal of Evolutionary Psychology,
7, 167–184. doi:10.1556/JEP.7.2009.2.5
Hall, D. L., Cohen, A. B., Meyer, K. K., Varley, A. H., &
Brewer, G. A. (2015). Costly signaling increases trust,
even across religious affiliations. Psychological Science,
26, 1368–1376. doi:10.1177/0956797615576473
Hayes, A. F. (2013). An introduction to mediation, modera-
tion, and conditional process analysis: A regression-based
approach. New York, NY: Guilford Press.
Hayes, A. F., & Preacher, K. J. (2014). Statistical mediation
analysis with a multicategorical independent variable.
British Journal of Mathematical and Statistical Psychology,
67, 451–470. doi:10.1111/bmsp.12028
Jonason, P. K., Li, N. P., & Webster, G. D. (2009). The dark triad:
Traits that facilitate short-term mating in men. European
Journal of Personality, 23, 5–18. doi:10.1002/per.698
Landolt, M. A., Lalumière, M. L., & Quinsey, V. L. (1995). Sex
differences in intra-sex variations in human mating tactics:
An evolutionary approach. Ethology and Sociobiology, 16,
3–23.
Litman, L., Robinson, J., & Abberbock, T. (2017). TurkPrime
.com: A versatile crowdsourcing data acquisition platform
for the behavioral sciences. Behavior Research Methods,
49, 433–442. doi:10.3758/s13428-016-0727-z
McCullough, M. E., Swartwout, P., Carter, E. C., Shaver, J. H.,
& Sosis, R. (2016). Christian religious badges instill trust
in Christian and non-Christian perceivers. Psychology of
Religion and Spirituality, 8, 149–163. doi:10.1037/rel0000045
McCullough, M. E., & Willoughby, B. L. B. (2009). Religion,
self-regulation, and self-control: Associations, explana-
tions, and implications. Psychological Bulletin, 135, 69–
93. doi:10.1037/a0014213
Neel, R., Neufeld, S. L., & Neuberg, S. L. (2013). Would an obese
person whistle Vivaldi? Targets of prejudice self-present
to minimize appearance of specific threats. Psychological
Science, 24, 678–687. doi:10.1177/0956797612458807
Norenzayan, A., Shariff, A. F., Willard, A. K., Slingerland, E.,
Gervais, W. M., McNamara, R. A., & Henrich, J. (2016).
The cultural evolution of prosocial religions. Behavioral &
Brain Sciences, 39, Article e1. doi:10.1017/S0140525X1400
1356
Penke, L., & Asendorpf, J. B. (2008). Beyond global sociosex-
ual orientations: A more differentiated look at sociosexual-
ity and its effects on courtship and romantic relationships.
Journal of Personality and Social Psychology, 95, 1113–
1135. doi:10.1037/0022-3514.95.5.1113
Petersen, M. B., & Aarøe, L. (2015). Birth weight and social
trust in adulthood: Evidence for early calibration of
social cognition. Psychological Science, 26, 1681–1692.
doi:10.1177/0956797615595622
Pirlott, A. G., & MacKinnon, D. P. (2016). Design approaches
to experimental mediation. Journal of Experimental Social
Psychology, 66, 29–38. doi:10.1016/j.jesp.2015.09.012
Purzycki, B. G., Apicella, C., Atkinson, Q. D., Cohen, E.,
McNamara, R. A., Willard, A. K., . . . Henrich, J. (2016).
Moralistic gods, supernatural punishment and the
expansion of human sociality. Nature, 530, 327–330.
doi:10.1038/nature16980
Richerson, P. J., & Boyd, R. (2005). Not by genes alone: How
culture transformed human evolution. London, England:
University of Chicago Press.
Roes, F. L., & Raymond, M. (2003). Belief in moralizing
gods. Evolution and Human Behavior, 24, 126–135.
doi:10.1016/S1090-5138(02)00134-4
Rowthorn, R. (2011). Religion, fertility and genes: A dual
inheritance model. Proceedings of the Royal Society
B: Biological Sciences, 278, 2519–2527. doi:10.1098/
rspb.2010.2504
Schmitt, D. P., & Fuller, R. C. (2015). On the varieties of
sexual experience: Cross-cultural links between religiosity
and human mating strategies. Psychology of Religion and
Spirituality, 7, 314–326. doi:10.1037/rel0000036
Shariff, A. F., Willard, A. K., Andersen, T., & Norenzayan, A.
(2016). Religious priming: A meta-analysis with a focus

960 Moon et al.
on prosociality. Personality and Social Psychology Review,
20, 27–48. doi:10.1177/1088868314568811
Simpson, J. A. (2007). Psychological foundations of trust.
Current Directions in Psychological Science, 16, 264–268.
Stearns, S. C. (1992). The evolution of life histories. Oxford,
England: Oxford University Press.
Tan, J. H. W., & Vogel, C. (2008). Religion and trust: An
experimental study. Journal of Economic Psychology, 29,
832–848. doi:10.1016/j.joep.2008.03.002
Weeden, J., Cohen, A. B., & Kenrick, D. T. (2008). Religious
attendance as reproductive support. Evolution and
Human Behavior, 29, 327–334. doi:10.1016/j.evolhum
behav.2008.03.004
Weeden, J., & Kurzban, R. (2013). What predicts religiosity?
A multinational analysis of reproductive and cooperative
morals. Evolution and Human Behavior, 34, 440–445.
doi:10.1016/j.evolhumbehav.2013.08.006
Williams, K. E. G., Sng, O., & Neuberg, S. L. (2016). Ecology-
driven stereotypes override race stereotypes. Proceedings
of the National Academy of Sciences, USA, 113, 310–315.
doi:10.1073/pnas.1519401113
Wilson, D. S., O’Brien, D. T., & Sesma, A. (2009). Human pro-
sociality from an evolutionary perspective: Variation and
correlations at a city-wide scale. Evolution and Human
Behavior, 30, 190–200. doi:10.1016/j.evolhumbehav.2008
.12.002