Unfounded beliefs among teachers: The interactive role of rationality priming and cognitive ability

Full text

Unfounded beliefs among teachers: The interactive role of rationality
priming and cognitive ability
Jais Troian*
Aix Marseille Univ., LPS, Aix-en-Provence, France
Denis Caroti
Aix Marseille Univ., Centre Gilles Gaston Granger, CNRS, Marseille, France
Aix-Marseille Univ., CORTECS team, Marseille, Marseille, France
Thomas Arciszewski
Aix Marseille Univ., Centre Psycle, Aix-en-Provence, France
Tomas Ståhl
University of Illinois at Chicago, Psychology Department, Chicago (IL), USA
Abstract
Previous research suggests that Unfounded Beliefs (UB) -- such as conspiracist beliefs and
beliefs in the supernatural -- stem from similar cognitive and motivational mechanisms. More
specifically, it has been demonstrated that cognitive ability is negatively associated with UB,
but only among individuals who value epistemic rationality. The present study goes beyond
previous correlational studies by examining whether the negative association between
cognitive ability and UB can be strengthened through a subtle rationality prime. In a large
scale online experiment (N = 762 French teachers), we demonstrate that priming rationality
(vs. control) does enhance the negative relationship between cognitive ability and adherence
to supernatural beliefs, as well as Conspiracy Mentality (d = .2). This effect was not obtained
for Illusory Pattern Perception. This study’s usefulness as a ‘proof of concept’ for future
interventions aimed at reducing UB prevalence among the general public is discussed.
KEYWORDS: conspiracy mentality, supernatural beliefs, pattern perception, cognitive
ability, epistemic rationality
* jadam@aus.edu

INTRODUCTION
Individuals adhere to all sorts of unfounded beliefs (UB, i.e. beliefs that are not
warranted based on the available evidence). For instance, polls conducted in both America and
Europe have found that 37% of US citizens believe in haunted houses, as do 40% of UK citizens
and 28% of Canadians (Gallup, 2005). In that same poll, 21% of Americans and 13% of
Canadians/Britons declared they believed in witches. A more recent survey of beliefs in the US
found that at least half of the population believed in the existence of ancient civilizations like
Atlantis, and a quarter of them believed either that certain individuals have the ability to move
objects with their mind, or that aliens visited Earth in ancient times (Chapman University,
2017). Likewise, many reports point at the existence of a substantial prevalence of conspiracist
beliefs in both the US and Europe (see Ståhl & Van Prooijen, 2018). In France for instance,
22% of a nationally representative sample completely agreed with the statement that ‘the
government does not really govern and that we do not know who really pulls the strings’
(Gombin, 2013).
Far from being innocuous beliefs about the world we live in, UB can have negative
consequences. Exposure to conspiracist beliefs has been shown to decrease voting intentions
and the will to reduce one’s carbon footprint (Jolley & Douglas, 2014a), as well as willingness
to vaccinate one’s children (Jolley & Douglas, 2014b). Seemingly harmless paranormal beliefs
are linked to a preference for alternative medicine (Van den Bulck, & Custers, 2009), which in
turn is predictive of increased mortality rates among cancer patients (Johnson, Park, Gross, &
Yu, 2018). Notably, different forms of UB, such as supernatural beliefs and belief in various
conspiracies are highly correlated (e.g. Darwin, Neave, & Holmes, 2011), and associated with
lower acceptance of scientific knowledge (Lewandowsky, Gignac, & Oberauer, 2013). UB can
therefore be conceived of as a cluster of self-reinforcing beliefs that have various harmful
physical and social consequences. Consequently, there is currently a surge in research

investigating the psychological underpinnings of UB, and how such beliefs are best prevented
from spreading in society.
Research has identified three main classes of common antecedents of UB: motivational,
personality-related and cognitive (Ståhl & van Prooijen, 2018). Three types of motivations
underlie adherence to UB, namely (a) epistemic (b) existential (the need to feel safe and in
control) and (c) social (the need to belong with a group, see Douglas, Sutton & Cichocka, 2017
for an overview). Empirical findings support this classification, because adherence to UB is
positively associated with uncertainty reduction (Marchlewska, Cichocka, & Kossowska,
2018), loss of control (Whitson & Galinsky, 2008), death anxiety (Newheiser, Farias, & Tausch,
2011), and social exclusion threats (Graeupner & Coman, 2017). UB have also been found to
decrease when individuals are primed to resist persuasion (Bonetto, Troian, Varet, Lo Monaco,
& Girandola, 2018). Personality traits such as low Agreeableness (Swami et al., 2011), high
schizotypy (Holm, 2009), and high paranoid ideation (Fenigstein & Vanable, 1992) are all
associated with conspiracist beliefs.
Additionally, various forms of UB stem from similar basic cognitive mechanisms, such
as the tendency to perceive patterns in random noise (see van Prooijen, Douglas, & De
Inocencio, 2017). UB are also negatively associated with analytical reasoning (Pennycook,
Cheyne, Seli, Koehler, & Fugelsang, 2012; Swami, Voracek, Stieger, Tran, & Furnham, 2014).
Analytical reasoning negatively predicts adherence to UB (e.g. Hergovich & Arendsay, 2005),
and mediates the negative relationship between education level and UB (Van Prooijen, 2017),
whereas intuitive reasoning seems to be positively linked with UB (Sunter, 2018). However,
more recent developments in the literature have shown that motivational factors play an
important role in determining the relationship between analytical reasoning and UB, to the point
where high analytical capabilities can backfire and produce even higher rates of UB if
individuals are motivated to defend their worldviews (Kahan, Peters, Dawson, & Slovic, 2017).

This is presumably because motivation shapes information processing. Although cognitive
ability will determine the quality of information processing, motivation determines the goal of
information processing (e.g., accuracy vs. belief confirmation), as well as what information is
selected for processing in the first place (Kruglanski, 2013).
Building on this line of reasoning, Ståhl and Van Prooijen (2018) recently argued that,
in order to protect the individual from UB, a high cognitive ability needs to be complemented
with motivation to rely on logic and evidence when forming and evaluating one’s beliefs (i.e.,
motivation to be epistemically rational). In the absence of motivation to be epistemically
rational, one’s cognitive abilities are likely to serve other information processing goals (e.g.,
belief confirmation), or to remain disengaged altogether. To test this hypothesis, these
researchers measured stable individual differences in motivation to be epistemically rational,
using the validated Importance of Rationality Scale (Ståhl, Zaal, & Skitka, 2016). They
demonstrated that analytic cognitive style (Study 1), as well as cognitive ability (Study 2), were
negatively associated with UB, but only among individuals who scored high (vs. low) on the
importance of rationality scale (Ståhl & Van Prooijen, 2018).
Although these studies provided important new insights regarding the psychology of
UB, they are not without their limitations. First, the studies were correlational. As a
consequence, it is unclear whether motivation to be epistemically rational has a causal effect on
UB. Moreover, because these studies relied on stable individual differences in motivation to be
epistemically rational, it remains unknown whether UB can be reduced through interventions
that target people’s current motivational state. The present investigation therefore aimed at
experimentally replicating the results obtained by Ståhl & Van Prooijen (2018) by testing the
hypothesis that a subtle manipulation of rationality motives salience – i.e. the simple motivation
to use logical reasoning - moderate the effects of general cognitive ability upon adherence to
UB, by using an experimental design and a prime pertaining to rationality (i.e. not necessarily

framed as a motivation to be rational). More specifically, we hypothesized that general
cognitive ability should be negatively associated with UB (conspiracist, supernatural and
illusory pattern perception), but that this association would become substantially stronger under
rationality priming. If this hypothesis received support, it would provide further corroboration
for the motivated cognition approach to UB and a ‘proof-of-concept’ mechanism that could be
subsequently used to design interventions and training programs aiming to reduce the
prevalence of UB among targeted populations. In fact, should the results corroborate our
hypothesis, they would point at the need to integrate a motivational component in training
sessions to be worked upon with participants (i.e. interventions should provide both analytical
skills and incite participants to use them in daily life).
The study was conducted in accordance with the 1964 Helsinki declaration (WMO,
1964) and its later amendments, the ethical principles of the French Code of Ethics for
Psychologists (CNCDP, 2012), and the 2016 APA Ethical Principles of Psychologists and Code
of Conduct (APA, 2017). Data underlying these findings are openly available at [OSF LINK
MASKED FOR ANONYMOUS PEER REVIEW].
METHODS
Design
The experiment was based on a simple two conditions between-subjects design (control
vs. rationality prime). To prime rationality, we decided to use a modified version of Bonetto et
al.’s (2018) paradigm, in which the concept of interest is primed by asking participants to
answer some scale items. This technique has been shown to yield replicable effects on
motivation as well as self-perception (Uhlmann & Cohen, 2007; Ford, O’Hare, & Henderson,
2013; Bonetto et al., 2018). Participants in the experimental condition were thus asked to
answer a single 7-point Likert item designed to prime rationality at the beginning of the
questionnaire (‘To what extent do you feel rational’), with response options ranging from 1 (not

rational at all) to 5 (very rational). Participants in the control condition were not presented with
any prime item prior to filling out the questionnaire.
Prior to data collection we conducted a power analysis with GPower (Faul, Erdfelder,
Buchner, & Lang, 2009) to determine the appropriate sample size. Given the minimal nature of
our prime, we decided to set the expected interaction effect size to d = .2 or r² = .01 (the smallest
effect size of interest according to Cohen’s 1988 classification). It revealed that 550 subjects
(275 per cell) were needed to detect an increase of r² = .01 with 80% power at α = .05 with 3
predictors (general cognitive ability, priming condition and their interaction) in a regression
model. Because of concerns regarding potential missing data, we aimed to recruit at least 300
participants in each condition (N = 600). This sample size would also allow for generating stable
estimates of the link between cognitive ability and UB in both conditions (see Schönbrodt, &
Perugini, 2013).
Participants & Recruitment Procedure
Our target population was French secondary school (including vocational) teachers from
the Provence-Alpes-Côtes-d’Azur (PACA) area, for two main reasons. First, this study was part
of a larger research program aiming to assess prevalence of UB among teachers and pupils, as
well as to test the efficacy of critical thinking training programs in reducing UB among these
two populations. Second, French teachers need a masters’ degree level qualification from the
same public institution (ESPE) in order to work. That means error variance should be reduced
in our experiment because education level was held constant in the present sample.
Computerized questionnaires were sent through the PACA National Education internal
computer system, disseminated via electronic mail to all teachers in the area (N = 13,488). Our
final sample consisted of 762 teachers (30.7% male, 8.7% unspecified, Mage = 43.53, SD = 9.34,
Myears-experience = 16.84, SD = 9.49), randomly assigned to one of the two experimental conditions.

Materials
The study was introduced as a study on personality, cognitive abilities and personal
worldviews. After a brief introduction section (which included the prime item in the
experimental condition) participants were invited to complete a series of tasks and measures as
listed below. Descriptive statistics for each measure can be found in table 1.
Illusory Pattern Perception task. We developed a measure of pattern perception similar
to that from Van Prooijen, Douglas, and De Inocencio (2017) by using the website
https://www.random.org. The only difference is that we used series of dices instead of coin
tosses. The task consisted in rating, for each of 10 series of 10 consecutive dice throws, the
extent to which those were completely random or completely determined. An 11th measure was
added to those by telling participants that those 10 series were in fact 100 throws with the same
dice, and asking them to rate the extent to which the results were random or determined (7-
point Likert, from 1 ‘completely random’ to 7 ‘completely determined’, M = 2.33, SD = 1.27, α
= .92).
General Cognitive Ability. General cognitive ability was measured with the same
Numeracy test (Schwartz, Woloshin, Black, & Welch, 1997), and similar cognitive reflection
items (CRT, Primi, Morsanyi, Chiesi, Donati, & Hamilton, 2016) as the ones used by Ståhl &
Van Prooijen (2018, Study 2). Notably, because we were unable to find a validated French
version of the verbal ability test used by Ståhl and Van Prooijen (2018), we decided not to
include this measure.
In total participants answered 9 cognitive ability items, of which 3 were taken from the
numeracy test, and 6 were CRT items (%correct = 65.89, SD = 25.42, α = .76).
Conspiracist Mentality. We then asked participants to fill out a French version of the
Conspiracist Mentality Questionnaire (Lantian, Muller, Nurra, & Douglas, 2016). It consists of

a series of 5 items for which participants have to assess the likelihood of veracity, and taps into
a general conspiracist mindset (see (Bruder, Haffke, Neave, Nouripanah, & Imhoff, 2013; 11-
point Likert, from 0% ‘completely unlikely’ to 100% ‘completely likely’, M = 57.14, SD = 18.73,
α = .82)
Beliefs in the supernatural. We finally asked participants to fill out a 7-item scale of
supernatural beliefs, which we created by taking one item related to each of the 7 dimensions
of supernatural beliefs from Bouvet, Djeriouat, Goutaudier, Py, and Chabrol (2014), which
included the following: ‘The soul keeps existing after physical death’, ‘Psychokinesis, i.e. the
ability to to move objects with one’s mental force, is real’, ‘There exist real cases of witchcraft’,
‘The number 13 brings bad luck’, ‘In specific states, such as sleeping or trance, the mind can
detach itself from the body’, ‘Astrology is a valid means to tell the future’. These items were
chosen on the basis of their saturation levels on Bouvet et al. (2014) factor’ analysis. We also
included a modified version of the original item pertaining to UB in the existence of extra-
terrestrials; from ‘There exist extraterrestrials on other planets’ (which might be statistically
likely given the infinite size of our universe) to ‘Extraterrestrials have already visited planet
earth’ (which is completely unfounded). Items were answered on a 7-point scale, ranging from
1 ‘totally disagree’ to 7 ‘totally agree’ (M = 1.90, SD = 1.02, α = .79).
Demographics. Participants were asked to indicate their gender, age, number of years
serving as a teacher, type of school (primary, secondary, vocational) and topic of teaching.
[INSERT TABLE 1 HERE]
RESULTS
Randomization checks
There were no substantial between group differences in sample size (Nprime = 377; Ncontrol
= 385); χ²(1) = .08, p = .77; age, t(695) = .40, p = .69; years of experience, t(690) = 1.43, p =

.15; topic of teaching, χ²(26) = 25.33, p = .50; or type of school χ²(5) = 1.78, p = .88. However,
groups were slightly unequal in terms of gender (%maleprime = 36.9; %malecontrol = 24.7), χ²(2)
= 15.22, p < .001. To rule out that gender is responsible for any effects obtained, we therefore
report analyses with and without gender as a covariate.
Correlation analyses
Zero-order correlations between all (standardized) variables can be seen in table 2.
Though mostly of small size, the correlations replicate what is typically found in the literature:
cognitive ability was negatively associated with all UB measures, and Illusory Pattern
Perception, whereas UB and Illusory Pattern Perception measures were all positively correlated
(and the association between Supernatural Beliefs and Conspiracy Mentality was the strongest).
[INSERT TABLE 2 HERE]
Hypothesis test
A t-test revealed a slight difference in Cognitive Ability between the experimental
conditions, t(760) = 2.15, p = .032, d = .16. Given the small size of this effect, the assumption
of independence between independent variables in regression analysis was not violated.
Consequently, moderation models were computed with the help of PROCESS (Model 1,
bootstrap Ntrials = 5000; Hayes, 2012) for each of our three dependent variables. We included
Priming condition (dummy coded) as a categorical moderator, and General Cognitive Ability
as the independent variable. More information regarding bootstrap methods and PROCESS can
be obtained in Hayes (2017). As can be seen in Table 3, the predicted interaction effect was
found on Conspiracy Mentality and Supernatural Beliefs, but not on Illusory Pattern Perception.
These effects remained the same when introducing gender as a covariate (for Conspiracy
Mentality, β = -.20, 95%CI[-.34; -.05], Supernatural Beliefs β = -.20, 95%CI[-.35; -.06], and

Illusory Pattern Perception, β = -.09, 95%CI[-.23; .05]). Thus, gender differences cannot
account for the results obtained.
Cognitive ability only predicted lower Conspiracist Mentality in the priming (vs.
control) condition, β = .04, 95%CI[-.14; .06] vs. β = -.24, 95%CI[-.34; -.14]. Furthermore,
although cognitive ability was associated with weaker Supernatural beliefs in both conditions,
this association was stronger in the priming (vs. control) condition, β = -.13, 95%CI[-.24; -.03]
vs. β = -.34, 95%CI[-.45; -.25].
[INSERT TABLE 3 HERE]
DISCUSSION
The present study was inspired by previous research indicating that cognitive ability and
motivation to be epistemically rational interactively influence various unfounded beliefs (Ståhl
& Van Prooijen, 2018). Building on this work, we set out to examine whether a subtle
experimental manipulation of motivation to be rational is sufficient to strengthen the negative
relationship between cognitive ability and UB. Consistent with this idea, we were able to
demonstrate that Rationality Priming strengthens the negative relationship between Cognitive
Ability and unfounded supernatural beliefs, as well as general Conspiracy Mentality. As
anticipated, the predicted interaction effect was small (r² = .01 (or d = .2). This is presumably
in part because our priming procedure was minimal, which actually provided for a strong test
regarding our moderation hypothesis (Platt, 1964).
Whereas individual differences in cognitive ability and the rationality prime
interactively affected both of our measures of UB, we did not obtain the same effect on Illusory
Pattern Perception. One possibility is that the priming procedure was simply too subtle to
substantially affect this measure. However, we believe a more plausible explanation for this
null effect is that illusory pattern perception relies more heavily on bottom-up

processing/intuitive thinking than do supernatural beliefs and Conspiracy Mentality (Tversky
& Lieberman, 2018). Although such bottom-up processes contribute to UB as well (Van
Prooijen et al., 2017), they may be less affected by motivation to be epistemically rational.
Consistent with this interpretation, Ståhl & Van Prooijen (2018) did not find an interaction
between analytic thinking and individual differences in motivation to be epistemically rational
on illusory pattern perception.
These results open up interesting avenues for application. Previous studies have shown
that scepticism toward unfounded beliefs can be promoted by providing people with specific
counter-arguments. The downside of such interventions is that they only target specific beliefs.
By contrast, the present findings suggest that interventions against UB could be targeted more
generally at people’s motivation to be epistemically rational. Future studies should examine
alternative ways to increase motivation to be epistemically rational. For example, rationality
priming could be tapped into through personal/social identity processes; interventions could
target motivation to see oneself as rational, or to identify with a group that values rationality
(Hogg & Terry, 2000). This implies that other primes pertaining so self-categorization, and
social identification may be as efficient (e.g. Van Bavel & Cunnigham, 2009; Ford et al., 2013)
in triggering rationality motives as our procedure and could be implemented in trainings (see
Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007). In addition, it may also be possible
for practitioners to generate a perception of a descriptive norm of rationality among attendees
to a training designed to enhance analytical/critical thinking (see Cialdini & Trost, 1998), thus
increasing their motivation to use those skills, at least during the training (which could still be
efficient to prevent backfire effects).
Before concluding, it should be noted that some caveats remain with our current study.
Because the present study relied exclusively on self-reported outcomes, concerns regarding
potential effects of demand characteristics are warranted. In particular, it is possible that the

rationality prime did not promote more rational beliefs, but merely made participants more
inclined to report rational beliefs. However, if the prime merely served as a demand
characteristic, or induced social desirability concerns, then we would argue that we should have
expected a main effect of the rationality prime on UB, which was not observed here. Instead,
the rationality prime strengthened the negative association between cognitive ability and UB -
- consistent with what has been observed in previous correlational studies (Ståhl & Van
Prooijen, 2018). That said, confidence in our interpretation of the present results would be
bolstered by future studies using measures of UB that are not based on self-report (e.g. fake
news transmission in a paradigm similar to Allport and Postman’s 1947 rumour transmission
study), and in which individual differences in socially desirable responding are controlled for.
Another potential limitation is that this experiment was conducted online. Participants
in online experiments are sometimes less motivated, more distracted, and more heterogeneous
than those in a laboratory setting, which may increase noise in the data. It is possible that our
choice to conduct this study online led to smaller effects of our manipulation than would have
been obtained in a laboratory setting (although sometimes online and laboratory settings yield
similar effect sizes, see Paolacci, Chandler, & Ipeirotis, 2010). Replication studies in more
controlled settings are needed to determine whether the online administration used impacted
the present results.
As is the case with any study that does not rely on a representative sample, the present
study also has limited external validity. First and foremost, our sample consisted of educated
individuals (i.e. teachers) who were motivated to participate in the study. Future studies should
examine whether subtle rationality primes have similar effects on less educated individuals
outside of the teaching profession. Our sample was also predominantly female. Even though
gender differences in UB are small (e.g. Utinans, Ancane, Tobacyk, Boyraz, Livingston, &
Tobacyk, 2015), this fact nonetheless restrains the generalizability of our findings to some

degree. Importantly, however, the results remained the same when we statistically controlled
for participant gender.
In closing, our results provide further evidence that cognitive ability and motivation to
be rational interactively shape skepticism toward various unfounded beliefs. In particular, the
present findings suggest that the link between cognitive ability and scepticism toward UB can
be strengthened by providing people with a subtle rationality prime. We believe these findings
are not only of theoretical interest, but may have important practical implications as well. In
particular, we believe the present study constitutes a ‘proof of concept’ experiment upon which
future field research may draw to design and test more specific interventions aimed at reducing
UB among targeted populations such as pupils, students, or members of the general public.
Future research should aim to replicate and extend these findings, using motivational primes
that could easily be implemented by practitioners.
These findings may also be relevant for addressing so called ‘backfire’ effects in
response to interventions tailored to lower adherence to UB. When people are motivated to
defend their worldviews, having a high (vs. low) cognitive ability can increase (rather than
decrease) bias in reasoning (Kahan et al., 2017). The present results suggest that it may be
possible to prevent such effects by targeting individuals’ motivation to be rational more
generally. Thus, interventions and training programs focused on promoting critical thinking
may benefit greatly from going beyond the teaching of reasoning skills, by also targeting the
motivation to be epistemically rational. The present findings suggest that such a strategy could
increase the likelihood that individuals will recruit their cognitive abilities in the service of
seeking the truth, rather than in the service of defending their worldviews.
REFERENCES
Allport, G. W., & Postman, L. (1947). The psychology of rumor. Oxford, England: Henry Holt.

American Psychological Association. (2017). Ethical principles of psychologists and code of
conduct. Retrieved from http://www.apa.org/ethics/code/index.aspx
Bonetto, E., Troïan, J., Varet, F., Lo Monaco, G., & Girandola, F. (2018). Priming Resistance
to Persuasion decreases adherence to Conspiracy Theories. Social Influence, 1-12.
Bouvet, R., Djeriouat, H., Goutaudier, N., Py, J., & Chabrol, H. (2014). Validation française
de la Revised Paranormal Belief Scale. L’Encéphale, 40(4), 308‑314.
Bruder, M., Haffke, P., Neave, N., Nouripanah, N., & Imhoff, R. (2013). Measuring
individual differences in generic beliefs in conspiracy theories across cultures:
Conspiracy mentality questionnaire. Frontiers in Psychology, 4, 225.
Chapman University (2017). Paranormal America. Retrieved from
https://blogs.chapman.edu/wilkinson/2017/10/11/paranormal-america-2017/
Cialdini, R. B., & Trost, M. R. (1998). Social influence: Social norms, conformity and
compliance. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social
psychology (pp. 151-192). New York: McGraw-Hill.
Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences. New York, NY:
Routledge Academic.
Commission Nationale Consultative de Déontologie des Psychologues (2012). Code de
deontologie des psychologues (1996, Amended February, 2012). Retrieved from
http://www.cncdp.fr/index.php/code-de-deontologie/code-de-deontologie-2012
Darwin, H., Neave, N., & Holmes, J. (2011). Belief in conspiracy theories: The role of
paranormal belief, paranoid ideation and schizotypy. Personality and Individual
Differences, 50, 1289–1293.
Douglas, K. M., Sutton, R. M., & Cichocka, A. (2017). The psychology of conspiracy

theories. Current directions in psychological science, 26(6), 538-542.
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using
G*Power 3.1: Tests for correlation and regression analyses. Behavior Research
Methods, 41, 1149-1160.
Fenigstein, A., & Vanable, P. A. (1992). Paranoia and self-consciousness. Journal of
personality and social psychology, 62(1), 129.
Ford, J., O’Hare, D., & Henderson, R. (2013). Putting the “we” into teamwork: Effects of
priming personal or social identity on flight attendants’ perceptions of teamwork and
communication. Human factors, 55(3), 499-508.
Gallup US (2005). Paranormal Beliefs Come (Super)Naturally to Some. Retrieved from
https://news.gallup.com/poll/19558/paranormal-beliefs-come-supernaturally-
some.aspx
Graeupner, D., & Coman, A. (2017). The dark side of meaning-making: How social exclusion
leads to superstitious thinking. Journal of Experimental Social Psychology, 69, 218-
222.
Gombin, J. (2013). Conspiracy theories in France. Interim report. Retrieved from
Counterpoint website
:http://web.archive.org/web/20140308021114/http://counterpoint.uk.com/wpcontent/u
ploads/2013/05/Conspiracy-Theories-in-France-interim-report-3rd-May.pdf
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable
mediation, moderation, and conditional process modeling. Retrieved from:
http://www.afhayes.com/public/process2012.pdf
Hayes, A. F. (2017). Introduction to Mediation, Moderation, and Conditional Process
Analysis (2nd Ed.). New York: Guilford Press.

Hergovich, A., & Arendasy, M. (2005). Critical thinking ability and belief in the paranormal.
Personality and Individual Differences, 38, 1805–1812.
Hogg, M. A., & Terry, D. I. (2000). Social identity and self-categorization processes in
organizational contexts. Academy of management review, 25(1), 121-140.
Holm, N. (2009). Conspiracy theorizing surveillance: considering modalities of paranoia and
conspiracy in surveillance studies. Surveillance & Society, 7(1), 36-48.
Johnson, S. B., Park, H. S., Gross, C. P., & Yu, J. B. (2018). Use of alternative medicine for
cancer and its impact on survival. JNCI: Journal of the National Cancer
Institute, 110(1).
Jolley, D., & Douglas, K. M. (2014a). The social consequences of conspiracism: Exposure to
conspiracy theories decreases intentions to engage in politics and to reduce one’s carbon
footprint. British Journal of Psychology, 105, 35–56.
Jolley, D., & Douglas, K. M. (2014b). The effects of anti-vaccine conspiracy theories on
vaccination intentions. PLoS ONE, 9, e89177.
Kahan, D. M., Peters, E., Dawson, E. C., & Slovic, P. (2017). Motivated numeracy and
enlightened self-government. Behavioural Public Policy, 1(1), 54-86.
Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108, 480–498.
Kruglanski, A. W. (2013). Lay epistemics and human knowledge: Cognitive and motivational
bases. Berlin: Springer Science & Business Media.
Lantian, A., Muller, D., Nurra, C., & Douglas, K. M. (2016). Measuring belief in conspiracy
theories: Validation of a French and English single-item scale. International Review of
Social Psychology, 29, 1–14.
Lewandowsky, S., Gignac, G. E., & Oberauer, K. (2013). The role of conspiracist ideation
and worldviews in predicting rejection of science. PLoS ONE, 8, e75637.

Marchlewska, M., Cichocka, A., & Kossowska, M. (2018). Addicted to answers: Need for
cognitive closure and the endorsement of conspiracy beliefs. European Journal of
Social Psychology, 48(2), 109-117.
Newheiser, A.-K., Farias, M., & Tausch, N. (2011). The functional nature of conspiracy
beliefs: Examining the underpinnings of belief in the Da Vinci Code conspiracy.
Personality and Individual Differences, 51, 1007–1011.
Paolacci, G., Chandler, J., & Ipeirotis, P. G. (2010). Running experiments on Amazon
Mechanical Turk. Judgment and Decision Making, 5(5), 411-419.
Pennycook, G., Cheyne, J. A., Seli, P., Koehler, D. J., & Fugelsang, J. A. (2012). Analytic
cognitive style predicts religious and paranormal belief. Cognition, 123, 335–346.
Platt, J. R. (1964). Strong inference. science, 146(3642), 347-353.
Primi, C., Morsanyi, K., Chiesi, F., Donati, M. A., & Hamilton, J. (2016). The Development
and Testing of a New Version of the Cognitive Reflection Test Applying Item Response
Theory (IRT): A New Version of the Cognitive Reflection Test Applying IRT. Journal
of Behavioral Decision Making, 29(5), 453‑469
Schönbrodt, F. D., & Perugini, M. (2013). At what sample size do correlations
stabilize? Journal of Research in Personality, 47(5), 609-612.
Schwartz, L. M., Woloshin, S., Black, W. C., & Welch, H. G. (1997). The role of numeracy in
understanding the benefit of screening mammography. Annals of internal
medicine, 127(11), 966–972.
Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2007). The
constructive, destructive, and reconstructive power of social norms. Psychological
science, 18(5), 429-434.
Ståhl, T., & van Prooijen, J. W. (2018). Epistemic rationality: Skepticism toward unfounded

beliefs requires sufficient cognitive ability and motivation to be rational. Personality
and Individual Differences, 122, 155–163.
Ståhl, T., Zaal, M. P., & Skitka, L. J. (2016). Moralized rationality: Relying on logic and
evidence in the formation and evaluation of belief can be seen as a moral issue. PLoS
One, 11(11), e0166332.
Sunter, N. (2018, unpublished). How Rational versus Intuitive Thinking Styles predict
Paranormal Belief. Retrieved from https://e-space.mmu.ac.uk/621722/
Swami, V., Coles, R., Stieger, S., Pietschnig, J., Furnham, A., Rehim, S., & Voracek, M.
(2011). Conspiracist ideation in Britain and Austria: Evidence of a monological belief
system and associations between individual psychological differences and real‑world
and fictitious conspiracy theories. British Journal of Psychology, 102(3), 443-463.
Swami, V., Voracek, M., Stieger, S., Tran, U. S., & Furnham, A. (2014). Analytic thinking
reduces belief in conspiracy theories. Cognition, 133, 572–585.
Tversky, A., & Liberman, V. (2018). Critical Thinking: Statistical Reasoning and Intuitive
Judgment. Columbia University Press.
Uhlmann, E. L., & Cohen, G. L. (2007). “I think it, therefore it’s true”: Effects of self-
perceived objectivity on hiring discrimination. Organizational Behavior and Human
Decision Processes, 104, 207–223.
Utinans, A., Ancane, G., Tobacyk, J. J., Boyraz, G., Livingston, M. M., & Tobacyk, J. S.
(2015). Paranormal beliefs of latvian college students: a latvian version of the revised
paranormal belief scale. Psychological reports, 116(1), 116-126.
Van Bavel, J. J., & Cunningham, W. A. (2009). Self-categorization with a novel mixed-race

group moderates automatic social and racial biases. Personality and Social Psychology
Bulletin, 35(3), 321-335.
Van den Bulck, J., & Custers, K. (2009). Belief in complementary and alternative medicine is
related to age and paranormal beliefs in adults. European Journal of Public
Health, 20(2), 227-230.
Van Prooijen, J.-W. (2017). Why education predicts decreased belief in conspiracy theories.
Applied Cognitive Psychology, 31, 50–58.
Van Prooijen, J. W., Douglas, K. M., & De Inocencio, C. (2017). Connecting the dots:
Illusory pattern perception predicts belief in conspiracies and the supernatural.
European Journal of Social Psychology.
Whitson, J. A., & Galinsky, A. D. (2008). Lacking control increases illusory pattern
perception. Science, 322, 115–117.
World Medical Organization. (1964). Recommendations guiding physicians in biomedical
research involving human subjects. Helsinki.
TABLES
Table 1. Summary table of the descriptive statistics for Illusory Pattern Perception, Cognitive
Ability, Conspiracist Mentality and Supernatural Beliefs across experimental conditions
(N=762).
Characteristic
Control
(N = 385)
Priming
(N = 377)
Illusory Pattern perception
Cognitive Ability
Conspiracy Mentality
Supernatural Beliefs
2.42 (1.27)
0.70 (.24)
56.98 (19.03)
1.91 (1.01)
2.38 (1.33)
0.66 (.26)
57.29 (18.44)
1.88 (1.03)
Note. Number between brackets represent SDs.

Table 2. Summary of Correlation Analyses across conditions between, Cognitive Ability,
Illusory Pattern Perception, Conspiracy Mentality and Supernatural Beliefs (N = 730).
1
2
3
4
5
General Cognitive Ability
-
Illusory Pattern Perception
-.14***
-
Conspiracy Mentality
-.15***
.14***
-
Supernatural Beliefs
-.23***
.24***
.36***
-
Note. Numbers represent Pearson correlation coefficients. *** p < .001.
Table 3. Regression analyses associated with the moderation model (Outcomes: Model 1 =
Conspiracy Mentality, N = 742; Model 2 = Supernatural beliefs, N = 730; Model 3 = Illusory
Pattern Perception, N = 762).
Predictors
T
F
Df
b(s.e.)
95%CI
R²
P
Model 1
8.01***
(3, 738)
.03
< .001
GCA
1.33
.16(.12)
[-.07, .38]
.18
RP
.01
.01(.07)
[-.14, .14]
.99
GCA x RP
-2.73**
-.20(.07)
[-.34, -.05]
.01
.007
Model 2
18.84***
(3, 726)
.07
< .001
GCA
.69
.08(.12)
[-.15, .31]
.49
RP
-.68
-.05(.07)
[-.19, .09]
.49
GCA x RP
-2.92**
-.21(.07)
[-.36, -.07]
.01
.004
Model 3
6.11***
(3, 758)
.02
< .001
GCA
.09
.01(.11)
[-.21, .22]
.93
RP
-.84
-.06(.07)
[.19, .07]
.40
GCA x RP
-1.40
-.09(.07)
[-.23, .04]
.16
Note. RP = Rationality Priming, GCA = General Cognitive Ability, B = standardized
regression coefficients, s.e. = standard error. **p < .01, ***p < .001.

Figure 1. Moderation models for each dependent variable. Blue lines represent regression
slopes in the control condition, red lines in the rationality priming condition. 95%CI bounds
were computed for each slope.