Content uploaded by Jan Willem van Prooijen
Author content
All content in this area was uploaded by Jan Willem van Prooijen on Jan 23, 2018
Content may be subject to copyright.
The Influence of Control on Belief in Conspiracy Theories: Conceptual and Applied
Extensions
JAN-WILLEM VAN PROOIJEN
1
* and MICHELE ACKER
2
1
Netherlands Institute for the Study of Crime and Law Enforcement; and VU University Amsterdam, Amsterdam, the Netherlands
2
Otterbein University, Westerville, USA
Summary: Threats to control have been found to increase belief in conspiracy theories. We argue, however, that previous research
observing this effect was limited in two ways. First, previous research did not exclude the possibility that affirming control might
reduce conspiracy beliefs. Second, because of artificial lab procedures, previous findings provide little information about the ex-
ternal validity of the control threat–conspiracy belief relationship. In Study 1, we address the first limitation and find that affirming
control indeed reduces belief in conspiracy theories as compared with a neutral baseline condition. In Study 2, we address the sec-
ond limitation of the literature. In a large-scale US sample, we find that a societal threat to control, that citizens actually experi-
enced, predicts belief in a range of common conspiracy theories. Taken together, these findings increase insight in the
fundamental relationship between the human need for control and the tendency to believe in conspiracy theories. Copyright ©
2015 John Wiley & Sons, Ltd.
INTRODUCTION
Against the background of economic and financial crises,
global warming, wars, and epidemics, conspiracy theories
have widespread appeal. The Internet is filled with examples
of conspiracy theories, including allegations that 9–11 was an
inside job, that the US government was involved in the assas-
sination of John F. Kennedy, or that the US government has
been withholding evidence for the existence of intelligent
extraterrestrial life. Conspiracy theories can be defined as
explanatory beliefs, involving multiple actors who join
together in secret agreement and try to achieve a hidden goal
that is perceived as unlawful or malevolent (Bale, 2007; see
also Zonis & Joseph, 1994). Large numbers of ordinary
citizens believe in conspiracy theories, and hence, most con-
spiracy beliefs cannot be dismissed as pathological (Oliver &
Wood, 2014; Sunstein & Vermeule, 2009). Furthermore,
accumulating research findings reveal a range of detrimental
perceptions and behaviors that are associated with conspiracy
beliefs, including health problems, decreased civic virtue,
hostility, and radicalization (Inglehart, 1987; Jolley &
Douglas, 2014; Swami et al., 2011; Thorburn & Bogart,
2005; Van Prooijen, Krouwel, & Pollet, 2015). As a conse-
quence, there has been a surge in empirical research on this
phenomenon in recent years.
Although conspiracy theories vary enormously in content,it
has been noted that belief in such theories is grounded in sim-
ilar underlying psychological processes. For instance, a good
predictor of belief in one conspiracy theory is belief in a differ-
ent, conceptually unrelated conspiracy theory (e.g., Douglas &
Sutton, 2011; Goertzel, 1994; Lewandowski, Oberauer, &
Gignac, 2013; Swami, Chamorro-Premuzic, & Furnham,
2010; Swami et al., 2011, 2013; Van Prooijen et al., 2015;
Wood, Douglas, & Sutton, 2012). This suggests that people
vary in the extent to which they have a conspiratorial mindset,
prompting them to assume evil conspiracies as responsible for
impactful societal events. The main psychological process that
has been associated with this conspiratorial mindset is the
human desire to make sense of their social environment. Early
writings by Hofstadter (1966) already noted that conspiracy
beliefs help citizens who feel powerless or voiceless to
understand complex and distressing societal events (see also
Bale, 2007). Consistently, empirical findings suggest that the
motivation to make sense of events that potentially threaten
one’s community predicts belief in conspiracy theories
(Van Prooijen & Van Dijk, 2014; see also Van Prooijen &
Van Lange, 2014).
People’s desire to make sense of the social world is
closely coupled with the extent to which they experience
control over their environment. Various complementary the-
oretical perspectives, on meaning-making (Heine, Proulx, &
Vohs, 2006; Park, 2010; Van den Bos, 2009), paranoia
(Kramer, 1998), and compensatory control (Kay, Whitson,
Gaucher, & Galinsky, 2009; Rutjens, van Harreveld, & van
der Pligt, 2013), assume that threats to control increase
people’s mental efforts to make sense of the social world,
imbuing the world with meaning, purpose, and order. These
insights may explain why conspiracy theories seem to gain
momentum particularly following impactful societal events
that are likely experienced as control threats by citizens
(e.g., a terrorist strike, a war, or a natural disaster; see Pipes,
1997; Robins & Post, 1997; Shermer, 2011). Indeed,
research reveals that people are more likely to attribute
impactful, harmful societal events (e.g., a politician is assas-
sinated) to conspiracies than societal events that are less
impactful or harmful (e.g., someone tries to assassinate a
politician but fails; see McCauley & Jacques, 1979), a
finding that is attributable to people’s sense-making motiva-
tion (Van Prooijen & Van Dijk, 2014).
In a similar vein, various operationalizations of control
threats have been found to predict conspiracy beliefs. For in-
stance, an external locus of control—that is, a dispositional
tendency to believe that one’s outcomes are controlled by
external forces—is correlated with interpersonal mistrust
and paranoia (Mirowsky & Ross, 1983) and belief in
conspiracy theories (Hamsher, Geller, & Rotter, 1968).
*Correspondence to: Jan-Willem van Prooijen, Department of Social and
Organizational Psychology, VU University Amsterdam, Van der
Boechorststraat 1, 1081 BT Amsterdam, the Netherlands.
E-mail: j.w.van.prooijen@vu.nl
Copyright © 2015 John Wiley & Sons, Ltd.
Applied Cognitive Psychology,Appl. Cognit. Psychol. 29: 753–761 (2015)
Published online 10 August 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/acp.3161
Furthermore, a seminal study by Whitson and Galinsky
(2008) reveals that experimentally induced control threats
increases the extent to which participants perceive patterns,
such as images in random noise, patterns in stock market
information, and conspiracies. Complementary findings
indicate that control threats elicit responses that are widely
associated with conspiracy belief, such as attributing
increased power to one’s enemies (Sullivan, Landau, &
Rothschild, 2010), and scapegoating (Rothschild, Landau,
Sullivan, & Keefer, 2012). Furthermore, constructs that are
closely associated with control threats, such as death anxiety
(Newheiser, Farias, & Tausch, 2011), uncertainty (Van
Prooijen & Jostmann, 2013), and attitudinal ambivalence
(van Harreveld, Rutjens, Schneider, Nohlen, & Keskinis,
2014), have been found to similarly influence conspiracy
beliefs. In the following, we discuss how the present
contribution is designed to expand on these insights.
THE CURRENT RESEARCH
The previously discussed studies are important because they
empirically established a link between people’s desire for
control and people’s tendency to believe in conspiracy theo-
ries. Yet, we propose that the empirical evidence reviewed
earlier is limited in two ways. As a first limitation, empirical
research manipulating control in the context of belief in
conspiracy theories has typically used a design in which a
control threat condition was compared with a condition
where a sense of control was reaffirmed—for instance, by
asking participants to remember an event from their own life
in which they did have control—without including a neutral
base-rate condition (Sullivan et al., 2010; see also Whitson
& Galinsky, 2008). As such, the currently available empiri-
cal evidence does not exclude the alternative possibility that
reaffirming a sense of control actually reduces conspiracy
beliefs. Such an assertion would be in line with the same
theoretical perspectives that predict control threats to
increase conspiracy beliefs (Heine et al., 2006; Hofstadter,
1966; Kay et al., 2009; Kramer, 1998; Park, 2010): After
all, reaffirming a sense of control reduces the need to find
meaning in the environment through mental sense-making
processes, attenuating the potential for conspiracy beliefs.
Consistent with this possibility, it has been noted that the
trust that citizens have in politicians—who are frequently
implicated actors in conspiracy theories—is structurally
low and has been that way since at least the 1970s
(Andeweg, 2014). Furthermore, people often hold negative
stereotypes of politicians, portraying them as cold and
unreliable in general (Fiske & Durante, 2014). Although
these insights do not necessarily generalize to conspiracy
beliefs, they do suggest a structurally suspicious perception
of power holders among many citizens, also when control
is not threatened. Furthermore, a methodological point is that
people usually develop conspiracy beliefs in the context of
events, institutions, or social groups that people subjectively
perceive as threatening; hence, one might wonder how much
an experimentally induced control threat adds to the feeling
of distress people already experience when considering
how plausible various conspiracy theories are. To establish
whether affirmations of control might decrease conspiracy
beliefs, one needs a balanced design where high and low
control conditions are compared with a neutral base-rate
condition.
As a second limitation, most of the previously discussed
findings on control and conspiracy beliefs offer little infor-
mation about the external validity of this relationship. For
instance, conspiracy beliefs are frequently measured in the
context of fictitious scenarios or events (McCauley &
Jacques, 1979; Newheiser et al., 2011; van Harreveld et al.,
2014; Van Prooijen & Van Dijk, 2014; Whitson & Galinsky,
2008), and control threats are often operationalized by means
of an experimental priming procedure that people are
unlikely to encounter in everyday life (e.g., asking partici-
pants to write down an event where they had or lacked con-
trol; e.g., Rothschild et al., 2012; Sullivan et al., 2010; see
also Van Prooijen & Jostmann, 2013). Furthermore,
although other research questions in the psychology of
conspiracy beliefs have been tested in broader samples
(e.g., Swami et al., 2011, 2013; Van Prooijen et al., 2015),
most of the studies that focus specifically on the relation
between control and conspiracy beliefs are based on student
samples of convenience, raising questions about the genera-
lizability of the findings to a broader population (Henrich,
Heine, & Norenzayan, 2010). As such, theorizing on belief
in conspiracy theories can, and should be, meaningfully
extended by focusing on real-life control threats and their
impact on widely endorsed conspiracy theories among a
large and diverse sample of citizens.
In the present contribution, we conduct two studies that
address these issues. In Study 1, we manipulate control by
means of three conditions, namely a low control condition, a
high control condition, and a neutral base-rate condition.
Study 1 is hence designed to conceptually extend previous
research by establishing if the effects of experimentally
induced control manipulations on conspiracy beliefs are
attributable to the threat, or instead to the affirmation, of
control. Study 2 is a correlational study using a large-scale
and diverse sample in the USA. This study extends Study 1
and previous research by focusing on a real-life control threat
that people actually encountered, and its association with
belief in a range of common conspiracy theories among
ordinary citizens. This study is hence designed to test whether
previous experimental findings on the control–conspiracy
belief relationship will transfer to predict belief in conspiracy
theories in the context of a real-life control threat.
STUDY 1
In Study 1, we examine conspiracy beliefs about public
policy. It is often the case that citizens believe there has been
some unfair favor shown to certain businesses by politicians.
Belief in such pandering is related to the extent to which
people have or lack control (McGraw, Lodge, & Jones,
2002), suggesting that it may be a forerunner to a full-blown
conspiracy theory. The context for the first study was the
building of a new underground metro line for the city of
Amsterdam to connect the northern part of Amsterdam with
the southern part (the ‘North–south metro line’). This project
754 J.-W. van Prooijen and M. Acker
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
has been plagued by various problems, which were particu-
larly salient among Dutch citizens at the time we conducted
this study. It was over budget, behind schedule, and received
little support from the citizens of Amsterdam. Moreover, at
the time we collected the data of this study, the project was
frequently in the Dutch news as various old houses were
literally sinking into the ground due to damages to the
foundations caused by the underground constructions.
Consequentially, various residents had to be evacuated from
their houses.
In our study, we first primed participants with either high
or low control. In addition, we inserted a baseline condition
that was unrelated to control. Based on our line of reasoning,
and consistent with previous findings (e.g., Sullivan et al.,
2010; Whitson & Galinsky, 2008), we predict that partici-
pants primed with low control would endorse stronger
conspiracy beliefs about the Amsterdam metro line than
participants primed with high control. With our neutral
base-rate condition, we seek to find out whether this finding
is attributable to increased conspiracy belief in the low con-
trol condition, or rather, by decreased conspiracy belief in
the high control condition.
Method
Participants and design
We tested the impact of the control manipulation on cons-
piracy beliefs by inducing a design with three conditions
(control: high control vs. low control vs. baseline).
1
The
experiment had the form of a paper-and-pencil task. Partici-
pants were approached by a research assistant in the student
cafeterias of VU University Amsterdam and were asked to
participate in exchange for a candy bar. A total of 119 parti-
cipants were recruited (39 men, 80 women; M
age
= 21.31,
SD = 3.33).
Materials and procedure
Participants were informed that the questionnaire consisted of
two independent parts: a recall task and a study on attitudes
toward the North–south Amsterdam metro line. Participants
started with the recall task. This task entailed the manipula-
tion of control (e.g., Whitson & Galinsky, 2008). In the high
control condition, participants were asked to recall, and sub-
sequently describe, a situation in which ‘something happened
and they were in complete control of the situation’. In the low
control condition, participants received a similar instruction
pertaining to a situation in which ‘something happened and
they had completely no control over the situation’. In the
neutral baseline condition, participants were asked to recall
and describe what they had for dinner last night.
After this task, participants continued with the study on
attitudes toward the North–south Amsterdam metro line.
Participants first read an article with some factual infor-
mation about this project. Then, we measured participants’
belief in conspiracy theories by assessing their agreement
to the following nine statements (1 = strongly disagree,
7=strongly agree): ‘The city council did not act in the inter-
est of citizens when making a decision upon the North–south
metro line’,‘The city council transferred parts of the budget
to the bank accounts of others’,‘The city council found the
safety of the constructions less important than their agree-
ments with the construction companies’,‘The city council
knew that the safety of residents would be jeopardized, but
moved forward with this plan nevertheless’,‘Members of
the city council received money from construction compa-
nies to set this plan in motion’,‘Members of the city council
deliberately withheld information to avoid hampering the
construction of the North–south metro line’,‘Members of
the city council deliberately ignored criticism on the project,
in order not to harm previously made agreements’,‘The city
council acted with integrity during the construction of the
North–south metro line’(recoded), and ‘The city council
acted morally during the construction of the North–south
metro line’(recoded). These nine items were averaged into
a reliable conspiracy beliefs scale (α= .84). After completing
the questionnaire, participants were thanked and provided
with a candy bar of their choice.
Results
A one-way analysis of variance revealed a significant effect
of the control manipulation on participants’conspiracy
beliefs, F(2, 116) = 4.98, p= .008; η
2
= .08. According to a
Least Significant Difference (LSD) test, participants in the
low control condition reported significantly stronger cons-
piracy beliefs (M= 4.14, SD = 0.59) than participants in the
high control condition (M= 3.58, SD = 0.90), p= .002. The
observed difference between the high and low control condi-
tions is consistent with previous research and underscores
the robustness of this finding (e.g., Sullivan et al., 2010;
Whitson & Galinsky, 2008). Participants in the neutral
baseline condition (M= 3.94, SD = 0.89), then, did not differ
significantly from participants in the low control condition
(p= .27), but they had significantly stronger conspiracy
beliefs than participants in the high control condition
(p= .046). As such, the effects of the control manipulation
were attributable to a reduction in conspiracy beliefs in the
high control condition.
Discussion
Including a neutral baseline condition provides for a novel
interpretation of the influence of control on conspiracy be-
liefs: At least in our study, the effect of the control manipu-
lation is driven by the high control condition reducing
conspiracy beliefs, more so than by the low control condition
increasing conspiracy beliefs, as compared with a neutral
baseline condition. This conclusion hinges on the assump-
tion that our baseline condition is truly neutral, of course.
If anything, recalling what one had for dinner last night
might install some control, as it is unusual for healthy adults
in modern societies to not have control over what to eat. This
interpretation is unlikely, however, as the baseline condition
differed significantly from the high control condition. It thus
seems safe to assume that our baseline condition was suffi-
ciently neutral with respect to control.
1
The full design also included a manipulation of the morality of the Amsterdam
municipality. However, this manipulation exerted no effect on a manipulation
check asking whether participants considered the Amsterdam municipality a
moral organization (F<1; M= 4.42, SD = 1.10). We therefore dropped this
manipulation from the analyses.
Control and conspiracy belief 755
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
Of importance, these findings do not imply that control
threats are unrelated to conspiracy beliefs—after all, it is
well-known from everyday life observation, as well as
from empirical research (McCauley & Jacques, 1979; Van
Prooijen & Van Dijk, 2014), that threatening societal
events elicit more conspiracy theorizing than non-
threatening societal events. A more plausible interpretation
of these findings is that experimentally inducing a control
threat adds little above and beyond the distressed feelings
that people already experience in the context of the present
study. After all, participants responded to a questionnaire
that (i) reminded them of threatening societal events (in
the present study, the construction of the North–south
metrolineinAmsterdam)and(ii)askedthemtoconsider
the possibility that harmful conspiracies might exist. Be
that as it may, the main conclusion of Study 1 is that
affirming a sense of control can buffer people against con-
spiracy theorizing in the face of distressing societal events.
This insight adds to the broader question of how belief in
conspiracy theories can be attenuated.
STUDY 2
Study 1 conceptually extends previous research on feelings
of control and belief in conspiracy theories, by addressing
the first observed limitation in the research literature: That
is, previously it was unclear whether affirming control might
reduce conspiracy beliefs. Study 2 seeks to address the
second observed limitation in the literature, which is that
the external validity of the findings on control and belief in
conspiracy theories cannot be guaranteed based on previous
research (e.g., Sullivan et al., 2010; van Harreveld et al.,
2014; Van Prooijen & Jostmann, 2013; Whitson & Galinsky,
2008). In Study 2, we investigate belief in a range of com-
mon conspiracy theories in the context of a genuine societal
control threat rather than a manufactured one. Moreover, in
Study 2, we test the relation between control and conspiracy
beliefs in a large-scale sample of the US general population.
To test the influence of a genuine control threat on political
conspiracy beliefs, in Study 2, we reanalyzed a yet unpub-
lished dataset that was collected back in 1999. During that
time, many people were facing the potential harmful societal
event of the Y2K bug (also referred to as the millennium
bug), which was the possibility that there would be major
computer malfunctions when the year 2000 came around that
would affect everyone’s daily life. Much time, money, and
energy were spent on getting the government and the popula-
tion ready for the possible shutdown of all major computer
systems. There was constant news coverage of the potential
threat of the Y2K bug suggesting citizens could expect failures
of banking systems, power plants, and water and food
supplies. In the last 3 months of 1999, a study was undertaken
with a large US sample that assessed citizens’perceptions of
the Y2K bug, as well as various perceptions of the
government, and a belief in a range of popular conspiracy
beliefs (e.g., about the Kennedy assassination, about the
cover-up of evidence for the existence of extraterrestrial life,
and various others). Hence, the survey contained all the
relevant measures for the present purposes. The survey also
contained various additional measures that allowed us to statis-
tically control for relevant variables.
In the analyses, we controlled for a range of demographics
(gender, age, political orientation, and education level). Fur-
thermore, we also controlled for two conceptual variables in
the analyses. The first is participants’trust in the govern-
ment. Independent from the extent to which the Y2K bug
threatens people’s sense of control, people may differ in
the extent to which they trust the government; relatedly,
people may blame the government for the Y2K situation.
Hence, statistically accounting for trust in the government
is important to establish the unique relationship between
the experienced control threat over the Y2K bug and belief
in conspiracy theories.
Second, we included a measure of the extent to which
people believed the Y2K situation itself was caused by a
conspiracy. Although the Y2K bug was regarded mostly as
a technological problem at the time, one prominent conspir-
acy theory was that the Y2K situation was an evil scheme by
computer programmers and businesses to make money. Spe-
cifically, the allegation was that these computer program-
mers sought to get paid for solving the problem (or hoax)
that they deliberately created themselves. Ironically, this spe-
cific conspiracy theory is in all likelihood negatively related
with the perception of control threat caused by the Y2K bug:
After all, according to this conspiracy theory, the computer
programmers are ultimately responsible for causing harm,
not the bug itself. Nevertheless, it is important to statically
account for belief in a Y2K-conspiracy theory in the analy-
ses, to ensure that it is the experience of Y2K control threat,
and not Y2K conspiracy theorizing, that predicts a range of
conceptually unrelated conspiracy beliefs. We hypothesized
that variations in the extent to which people experience a
control threat due to the Y2K bug predicts belief in a range
of common conspiracy theories.
Method
Participants
There were originally 1423 participants. After examining the
data for potential duplicate cases and incomplete data, there
were 1256 valid participants (771 men, 479 women, and 6
unknown). Participants came from all over the USA; median
age was between 35 and 44 years old,
2
median education
level was college degree, and median household income
ranged between $40 000 and $59 000. Politically, partici-
pants were slightly conservative on average (M= 4.83,
SD = 1.60; on a scale ranging from 1, extremely liberal,to
7, extremely conservative).
Procedure
Participants were solicited from a range of sites on the Inter-
net including the online experiment pages of several psycho-
logical societies, university websites, Yahoo, and Y2K
relevant websites. Participants could also find the survey
2
Note that it is impossible to report a precise mean and standard deviation of
participants’age, as we measured age through ranges on a 7-point scale (i.e.,
1=under 18;2=18–24;3=25–34;4=35–44;5=45–54;6=55–64;7=65
and over). Nevertheless, here we do report these statistics for the age scale:
M= 4.31, SD = 1.29.
756 J.-W. van Prooijen and M. Acker
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
via a search engine or by learning the address in the media
(e.g., television news programs). After providing informed
consent, the participants completed an online survey that
asked questions about the Y2K crisis (or the millennium
bug) and about a wide variety of political and psychological
attitudes. All participants completed the survey at the height
of public awareness and interest between 31 October 1999
and 1 December 1999.
Measures
Control threat
We measured the degree of control threat that people per-
ceived from the impending year 2000 by assessing their re-
sponses to the following five items: ‘To what extent do
you think you can personally control negative consequences
due to the Y2K situation?’(1 = little control,5=much con-
trol) (recoded), ‘How many people in the United States will
be exposed to risks from the Y2K situation?’(1 = few people;
5=many people)‘To what extent do you believe that you are
personally at risk from the Y2K situation?’(1 = not at risk;
5=very much at risk), ‘To what extent does Y2K have the
potential to cause catastrophic death and destruction around
the world?’(1 = very low catastrophic potential,5=very
high catastrophic potential) and ‘How likely is Y2K to cause
consequences that could be fatal?’(1 = certain not to be fa-
tal;5=certain to be fatal). These five items were averaged
into the control threat scale (α= .71).
Trust in government
We measured trust with four items from the General Social
Survey (Davis & Smith, 1996) (1 = strongly disagree,
5=strongly agree): ‘Those we elect to public office usually
try to keep the promises they have made during the election’,
‘The people running the country don’t really care what hap-
pens to you’(recoded), ‘The people in Washington, D.C. are
out of touch with the rest of the country’(recoded), and ‘You
can generally trust the people who run our government to do
what is right’. These four items were averaged into a reliable
measure of trust in government (α= .79).
Belief in a Y2K conspiracy theory
We assessed the extent to which participants believed that
the Y2K situation itself was a conspiracy with the following
item: ‘How likely is it that the Y2K situation is just a scheme
by computer programmers and businesses to make money?’
(1 = very unlikely,5=very likely).
Conspiracy beliefs
Our dependent measures were items assessing participants’
belief in a total of five commonly known conspiracy theo-
ries. Participants were asked to indicate how likely each of
the following statements was true or false (1 = definitely
false,5=definitely true): ‘The Air Force is hiding evidence
that the United States has been visited by flying saucers
(UFOs)’,‘The Japanese are deliberately conspiring to
destroy the American economy’,‘The American government
deliberately put drugs into inner city communities’,‘Presi-
dent Kennedy was killed by an organized conspiracy, not a
lone gunman’, and ‘O.J. Simpson was framed for the
murders of his ex-wife and her friend’. It turned out that
these five conspiracy theories were only moderately inter-
correlated (.10 <rs<.48; see Table 1), and therefore, we
chose to examine each conspiracy theory separately.
Results
The means, standard deviations, and correlations between the
measured variables are displayed in Table 1. We analyzed the
results with hierarchical regression analyses in which various
demographics (i.e., gender, age, education level, and political
orientation) were inserted as control variables in Step 1. In
Step 2, we added trust in government, and belief in a Y2K
conspiracy theory, to the regression model. The crucial con-
trol threat variable was added to the model in Step 3.
The results of the hierarchical regression analyses are
displayed in Table 2.
3
As can be seen in the table, Step 1
was significant for all conspiracy theories. These effects were
predominantly driven by education level, as lower education
level was associated with stronger belief in four out of five
conspiracy theories (UFO, Japanese, drugs, and Kennedy).
Political orientation had inconsistent effects, as the political
left was more inclined to endorse the UFO and O.J. Simpson
conspiracy theories, and the political right was more inclined
to endorse the Japanese conspiracy theory.
Step 2 was also significant for all five conspiracy theories:
As displayed in Table 2, trust in government negatively
3
Because of missing values, mostly on the demographics (Step 1), the de-
grees of freedom deviated from the full sample on all conspiracy theories.
Differences in degrees of freedom between specific conspiracy theories are
due to missing values on the conspiracy beliefs in question.
Table 1. Means, standard deviations, and inter-correlations of the measures in Study 2
MSD 1 2 345678
1. Y2K control threat 3.57 0.84 —
2. Trust in government 2.24 0.78 .37*** —
3. Y2K conspiracy belief 1.33 0.83 .26*** .12*** —
4. UFO conspiracy belief 2.73 1.26 .16*** .18*** .07* —
5. Japanese conspiracy belief 2.06 0.96 .17*** .19*** .10*** .33*** —
6. Drugs conspiracy belief 2.21 1.22 .27*** .31*** .03 .40*** .40*** —
7. Kennedy conspiracy belief 3.44 1.21 .19*** .23*** .05 .47*** .25*** .43*** —
8. O.J. Simpson conspiracy belief 1.55 0.86 .00 .05 .09** .11*** .12*** .19*** .09** —
Note: N per correlation ranged from 1236 to 1254 participants. All variables were measured on five-point scales.
*p<.05;**p<.01;***p<.001, two-sided.
Control and conspiracy belief 757
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
predicted belief in four out of five conspiracy theories (the
only exception being the O.J. Simpson conspiracy theory),
and belief in a Y2K conspiracy theory positively predicted
belief in four out of five conspiracy theories (the only excep-
tion being the drugs conspiracy theory).
The crucial test of our Hypothesis was Step 3, in which
control threat was added to the regression model. As can
be seen in Table 2, this step was significant for four out of
five conspiracy theories (UFO, Japanese, drugs, and
Kennedy), and the regression coefficient was positive for
all of these four conspiracy theories. These results support
the hypotheses that a societal control threat—in this case,
due to the millennium bug—predicts belief in a range of
conceptually unrelated conspiracy theories.
The only exception was belief in the O.J. Simpson conspir-
acy theory, where Step 3 was nonsignificant. As can be seen
in Table 1, belief in this conspiracy theory was overall quite
low (M= 1.55, SD = 0.86). We suspect that this conspiracy
theory is specific to the US African-American community.
For instance, research suggests that particularly African-
Americans believed O.J. Simpson to be innocent (Graham,
Weiner, & Zucker, 1997). Furthermore, it has been noted that
African-Americans are quite susceptible to conspiracy
theories assuming a White plot designed to harm fellow
African-Americans (Crocker, Luhtanen, Broadnax, &
Blaine, 1999; see also Simmons & Parsons, 2005; Thorburn
& Bogart, 2005). It might thus be possible that control threat
predicts belief in an O.J. Simpson conspiracy theory only
among African-Americans. It was impossible to test this line
of reasoning in the present data, however, as we did not have a
sufficient number of African-Americans in our sample to
conduct meaningful analyses.
Discussion
These findings illuminate the applied implications of the
control–conspiracy belief relationship. Whereas previous
research investigated this relation using fictitious scenar-
ios, or artificial operationalizations of control (e.g.,
McCauley & Jacques, 1979; Rothschild et al., 2012;
Sullivan et al., 2010; Van Prooijen & Van Dijk, 2014;
Whitson & Galinsky, 2008), the present study suggests
that real-life threats to control can have meaningful and
substantial psychological consequences for citizens. Spe-
cifically, our data reveal that the extent to which the
millennium bug threatened people’s sense of control
during the last 3 months of 1999 predicted their belief
in a range of common, and conceptually unrelated, con-
spiracy theories.
Unlike previous studies (e.g., Douglas & Sutton, 2011;
Goertzel, 1994; Lewandowski et al., 2013; Swami et al.,
2011; Van Prooijen et al., 2015), the five conspiracy beliefs
under investigation were only moderately correlated in this
study. We suspect that this is due to the rather different
topics of conspiracy theorizing across the five items.
For the present purposes, the finding that the predicted effect
emerged for four out of five conspiracy theories despite
their relatively low inter-correlations only underscores the
robustness of the relationship between control threats and
conspiracy beliefs.
GENERAL DISCUSSION
The present findings fit into a broad research domain designed
to uncover the psychological processes underlying belief in
conspiracy theories. Specifically, the extent to which people
are able to exert control over their social environment is
closely coupled with their desire to make sense of this
environment (e.g., Park, 2010; van den Bos, 2009). Such
sense-making motivation is a central ingredient of belief in
conspiracy theories (Hofstadter, 1966; see also Bale, 2007;
Van Prooijen & Van Dijk, 2014). These complementary
insights stimulated a wealth of studies illuminating that
Table 2. Hierarchical regression results (Study 2)
Conspiracy theory
UFO cover-up Japanese conspiracy Drugs conspiracy Kennedy conspiracy O.J. Simpson
Step 1 βt(1168) βt(1165) βt(1163) βt(1164) βt(1165)
Gender .06 1.92 .01 0.35 .07 2.29* .07 2.35* .01 0.38
Age .13 4.46*** .02 0.54 .05 1.75
†
.06 2.06* .02 0.77
Education level .21 7.06*** .11 3.56*** .13 4.17*** .16 5.39*** .02 0.67
Political orientation .13 4.52*** .07 2.17* .01 0.43 .04 1.25 .09 2.96**
R
2
= .06 R
2
= .02 R
2
= .02 R
2
= .03 R
2
= .01
F
4, 1168
= 18.06*** F
4, 1165
= 4.83*** F
4, 1163
= 5.59*** F
4, 1164
= 9.73*** F
4, 1165
= 2.40*
Step 2 βt(1166) βt(1163) βt(1161) βt(1162) βt(1163)
Trust in government .21 7.34*** .20 6.57*** .34 11.74*** .23 7.90*** .01 0.47
Y2K conspiracy belief .09 3.10** .12 4.16*** .05 1.86
†
.09 3.21** .08 2.75**
ΔR
2
= .05 ΔR
2
= .05 ΔR
2
= .11 ΔR
2
= .06 ΔR
2
= .01
F
2, 1166
= 30.55*** F
2, 1163
= 28.65*** F
2, 1161
= 69.53*** F
2, 1162
= 34.98*** F
2, 1163
= 3.99*
Step 3 βt(1165) βt(1162) βt(1160) βt(1161) βt(1162)
Control threat .14 4.67*** .16 4.98*** .22 7.42*** .13 4.03*** .06 1.71
†
ΔR
2
= .02 ΔR
2
= .02 ΔR
2
= .04 ΔR
2
= .01 ΔR
2
= .002
F
1, 1165
= 21.83*** F
1, 1162
= 24.77*** F
1, 1160
= 54.98*** F
1, 1161
= 16.27*** F
1, 1162
= 2.94
†
Note: Degrees of freedom vary per conspiracy theory because of missing values.
†
p<.10;*p<.05;**p<.01;***p<.001, two-sided.
758 J.-W. van Prooijen and M. Acker
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
people are most likely to believe in conspiracy theories when
they lack control (Hamsher et al., 1968; Newheiser et al.,
2011; Sullivan et al., 2010; Van Prooijen & Jostmann,
2013; Whitson & Galinsky, 2008). The present research
was designed to extend these insights in two ways. First,
previous research typically investigated the impact of control
on belief in conspiracy theories with incomplete research
designs, raising questions about the additional possibility that
affirmations of control might attenuate conspiracy beliefs.
Study 1 of the current contribution addresses this issue and
indeed reveals a reduction in conspiracy beliefs when
people’s sense of control is affirmed. Second, previous
findings offered little indications of the external validity of
the relation between control and belief in conspiracy
theories. Our Study 2 findings, however, reveal that a real-life
control threat predicts a range of conspiracy beliefs among a
large number of citizens. Taken together, the studies
presented here conceptually extend insights on the relation
between control and belief in conspiracy theories.
Both studies independently hold a valuable insight for
theorizing on belief in conspiracy theories. One main
implication of Study 1 is that the human need for control is
intimately related with belief in conspiracy theories in both
directions of a control continuum. Although in our Study
1, the low control condition did not differ significantly from
the neutral baseline condition, it stands to reason that the
specific setting of the study was already threatening to
control, also in the neutral condition (i.e., the study being
situated in the context of controversial public policy, with
observable harmful effects for many citizens). Hence, the
relation between control and belief in conspiracy theories is
in all likelihood bidirectional: Whereas threats to control
might increase belief in conspiracy theories (e.g., McCauley
& Jacques, 1979; Van Prooijen & Van Dijk, 2014; Whitson
& Galinsky, 2008), our findings reveal that affirmations of
control might just as well decrease belief in conspiracy theo-
ries. This latter point has not been recognized before in this
research domain.
A main implication of Study 2 is that the relationship
between the human need for control and belief in conspir-
acy theories is not confined to the lab. In popularized
writings, it is frequently noted that conspiracy theories
gain momentum when society faces uncertain times, such
as economic crises, threats of terrorism, climate change,
and the like (e.g., Pipes, 1997; Shermer, 2011). Empirical
evidence had not yet established that a real-life control
threat indeed predicts belief inconspiracytheories.Our
reanalysis of the Y2K-bug data indicates that a genuine
control threat predicts people’s susceptibility to unrelated
conspiracy theories, such as belief in a UFO cover-up,
and the belief in a Kennedy conspiracy theory. Indeed,
these relations hold up even after controlling for trust in
the government, and the specific belief in a Y2K conspir-
acy theory. These findings underscore the external validity
of the control–conspiracy belief relation and suggest that
societal threats to control are associated with a generally
suspicious, conspiratorial mindset.
The present research implies that societal developments
may predict the waxing and waning of popular conspiracy
beliefs over time. Citizens frequently are faced with
threatening events or periods, such as intergroup conflict
and riots, economic and financial crises, and wars. Based
on the present research, it stands to reason that conspiracy
beliefs are particularly likely to flourish in such times of so-
cietal turmoil. Indeed, it might be speculated that some of
these conspiracy beliefs have the potential of further deterio-
rating these detrimental societal circumstances by
undermining the legitimacy of public governance. These
considerations suggest that, whenever society is facing a
substantial control threat, political leaders have an increased
responsibility to install a sense of trust among the public, and
to overtly display signs of moral behavior, particularly if
they want to avoid increases in perceived threat and the
virulent spread of conspiracy beliefs. Although most of the
time this will turn out to be easier said than done, some
factors that may help political leaders to increase their moral
authority are being transparent about their motivations for
important decisions (Brockner, DeWitt, Grover, & Reed,
1990; Folger & Martin, 1986), and closely adhering to
procedural justice principles (Leventhal, 1980), as these
factors are known to support feelings of control. It may also
be useful to consider how leaders can reinforce other areas
where control is not threatened in order to ameliorate
conspiratorial thinking.
Also other sources of evidence suggest that the relation-
ship between threatening societal events and belief in
conspiracy theories can have substantial macro-political
implications. For instance, it has been noted that collective
control threats in the form of societal crises—particularly if
preceded by a short-lived period of prosperity—predict the
rise of politically extremist regimes (Midlarsky, 2011). Em-
pirical findings indeed underscore that fearful socio-
economic circumstances are associated with politically
extremist ideologies (Van Prooijen, Krouwel, Boiten, &
Eendebak, 2015). These observations are relevant for the
current purposes, because political extremism has been
found to predict belief in conspiracy theories (Van Prooijen
et al., 2015; see also Inglehart, 1987). It is yet unclear what
the causal order is of the relation between belief in conspir-
acy theories and political extremism. What is clear, though,
is that both phenomena fuel detrimental societal develop-
ments, such as fear, hostility, and intergroup conflict. Socie-
tal threats to control hence can have a major impact on the
functioning of society, and this is at least partly due to the
processes described in this contribution.
Both studies reported herein have their limitations. In
keeping with previous studies, Study 1 focused on an
experimenter-designed control threat, in a sample of
undergraduate university students. A primary concern of
Study 1 is therefore the question how meaningful these
findings are to predict the influence of societal develop-
ments on belief in conspiracy theories. This main weak-
ness of Study 1 is one of the main strengths of Study 2,
however, which focused on a genuine control threat in a
large sample of US citizens. Study 2, then, was a cross-
sectional design that precludes conclusions about cause
and effect. Based on Study 2 alone, we cannot be certain
whether variations in control caused belief in conspiracy
theories or instead that people with a conspiratorial
mindset were more likely to perceive the Y2K situation
Control and conspiracy belief 759
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
as a threat to control. This limitation, in turn, is addressed
by the main strength of Study 1, showing evidence for a
causal influence of variations in control on belief in con-
spiracy theories (cf. Sullivan et al., 2010; Whitson &
Galinsky, 2008). Although this finding still leaves open
the possibility of reverse-causation (i.e., a conspiratorial
mindset might also contribute to the experience of control
threat), it does install confidence in our interpretation of
the findings observed in the studies presented here.
To conclude, the present studies add to an emerging body
of literature on the psychological underpinnings of belief in
conspiracy theories. Whereas it is well-established that large
portions of ordinary citizens believe in conspiracy theories
(Oliver & Wood, 2014), there is much that can still be
learned about the question of why they do. The present stud-
ies sought to contribute to this issue by highlighting how
having or lacking control influences the sense-making pro-
cesses leading up to conspiracy beliefs. In doing so, the stud-
ies conceptually refine previous insights on the relationship
between control and conspiracy beliefs. It is concluded that
the human need for control is closely coupled with their ten-
dency to believe in conspiracy theories, a relationship that
can also be observed in applied settings.
ACKNOWLEDGEMENTS
We thank Anlacan Tran, Jarik Bouw, and Ruben Teitler for
their assistance in collecting the data of Study 1.
REFERENCES
Andeweg, R. B. (2014). A growing confidence gap in politics? Data versus
discourse. In J.-W. van Prooijen, & P. A. M. van Lange (Eds.), Power,
politics, and paranoia: Why people are suspicious of their leaders
(pp. 176–198). Cambridge, UK: Cambridge University Press.
Bale, J. M. (2007). Political paranoia v. political realism: On distinguishing
between bogus conspiracy theories and genuine conspiratorial politics.
Patterns of Prejudice,41,45–60.
Brockner, J., DeWitt, R.L, Grover, S., & Reed, T. (1990). When it is espe-
cially important to explain why: Factors affecting the relationship be-
tween managers’explanations of a layoff and survivors’reactions to
the layoff. Journal of Experimental Social Psychology,26, 389–407.
Crocker, J., Luhtanen, R., Broadnax, S., & Blaine, B. E. (1999). Belief in U.
S. government conspiracies against blacks among black and white college
students: Powerlessness or system blame? Personality and Social
Psychology Bulletin,25, 941–953
Davis, J. A., & Smith, T. W. (1996). General Social Surveys 1972–96:
Cumulative codebook. Chicago: National Opinion Research Center.
Douglas, K. M., & Sutton, R. M. (2011). Does it take one to know one? En-
dorsement of conspiracy theories is influenced by personal willingness to
conspire. British Journal of Social Psychology,50, 193–364.
Fiske, S. T., & Durante, F. (2014). Never trust a politician? Collective dis-
trust, relational accountability, and voter response. In J.-W. van Prooijen,
& P. A. M. van Lange (Eds.), Power, politics, and paranoia: Why people
are suspicious of their leaders (pp. 91–105). Cambridge, UK: Cambridge
University Press.
Folger, R., & Martin, C. (1986). Relative deprivation and referent
cognitions: Distributive and procedural justice effects. Journal of
Experimental Social Psychology,22, 531–546.
Goertzel, T. (1994). Belief in conspiracy theories. Political Psychology,15,
733–744.
Graham, S., Weiner, B. & Zucker, G.S. (1997). An attributional analysis of
punishment goals and public reactions to O. J. Simpson. Personality and
Social Psychology Bulletin,23, 331–346.
Hamsher, J. H., Geller, J. D., & Rotter, J. B. (1968). Interpersonal trust,
internal-external control, and the Warren Commission Report. Journal
of Personality and Social Psychology,9, 210–215.
Heine, S. J., Proulx, T., & Vohs, K. D. (2006). The meaning maintenance
model: On the coherence of social motivations. Personality and Social
Psychology Review,10,88–110.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in
the world? Behavioral and Brain Sciences,33,61–135.
Hofstadter, R. (1966). The paranoid style in American politics. In R.
Hofstadter (Ed.), The paranoid style in American politics and other es-
says (pp. 3–40). New York, NY: Knopf.
Inglehart, R. (1987). Extremist political position and perceptions of conspir-
acy: Even paranoids have real enemies. In C. F. Graumann, & S.
Moscovici (Eds.), Changing conceptions of conspiracy (pp. 231–244).
New York, NY: Springer-Verlag.
Jolley, D., & Douglas, K. (2014). The social consequences of conspiracism:
Exposure to conspiracy theories decreases intentions to engage in politics
and to reduce one’s carbon footprints. British Journal of Psychology,
105,35–56.
Kay, A. C., Whitson, J. A., Gaucher, D., & Galinsky, A. D. (2009).
Compensatory control: Achieving order through the mind, our institutions,
and the heavens. Current Directions in Psychological Science,18,264–268.
Kramer, R. M. (1998). Paranoid cognition in social systems: Thinking and
acting in the shadow of doubt. Personality and Social Psychology Re-
view,2, 251–275.
Leventhal, G. S. (1980). What should be done with equity theory?: New ap-
proaches to the study of fairness in social relationships. In K. J. Gergen,
M. S. Greenberg, & R. H. Willis (Eds.), Social exchange: Advances on
theory and research (pp. 27–54). New York: Plenum.
Lewandowski, S., Oberauer, K., & Gignac, G. (2013). NASA faked the
moon landing—Therefore (climate) science is a hoax: An anatomy of
the motivated rejection of science. Psychological Science,24, 622–633.
McCauley, C., & Jacques, S. (1979). The popularity of conspiracy theories
of presidential assassination: A Bayesian analysis. Journal of Personality
and Social Psychology,37, 637–644.
McGraw, K. M., Lodge, M., & Jones, J. M. (2002). The pandering politi-
cians of suspicious minds. Journal of Politics,64, 362–383.
Midlarsky, M. L. (2011). Origins of political extremism. Cambridge, UK:
Cambridge University Press.
Mirowsky, J., & Ross, C. E. (1983). Paranoia and the structure of powerless-
ness. American Sociological Review,48, 228–239.
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.
Oliver, J. E., & Wood, T. (2014). Medical conspiracy theories and health
behaviors in the United States. JAMA Internal Medicine,174, 817–818.
Park, C. L. (2010). Making sense of the meaning literature: An integrative
review of meaning making and its effects on adjustment to stressful life
events. Psychological Bulletin,136, 257–301.
Pipes, D. (1997). Conspiracy: How the paranoid style flourishes and where
it comes from. New York, NY: Simon & Schusters.
Robins, R. S., & Post, J. M. (1997). Political paranoia: The psychopolitics
of hatred. New Haven, CT: Yale University Press.
Rothschild, Z. K., Landau, M. J., Sullivan, D., & Keefer, L. A. (2012). A
dual-motive model of the motives underlying scapegoating: Displacing
blame to reduce guilt or increase control. Journal of Personality and So-
cial Psychology,102, 1148–1163.
Rutjens, B. T., Van Harreveld, F., & Van der Pligt, J. (2013). Step by step:
Finding compensatory order in science. Current Directions in Psycholog-
ical Science,22, 250–255.
Shermer, M. (2011). The believing brain: From ghosts and gods to politics
and conspiracies—How we construct beliefs and reinforce them as
truths. New York, NY: Henry Holt.
Simmons, W. P., & Parsons, S. (2005). Beliefs in conspiracy theories among
African Americans: A comparison of elites and masses. Social Science
Quarterly,86, 582–598.
Sullivan, D., Landau, M. J., & Rothschild, Z. K. (2010). An existential func-
tion of enemyship: Evidence that people attribute influence to personal
and political enemies to compensate for threats to control. Journal of
Personality and Social Psychology,98, 434–449.
760 J.-W. van Prooijen and M. Acker
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
Sunstein, C. R., & Vermeule, A. (2009). Conspiracy theories: Causes and
cures. The Journal of Political Philosophy,17, 202–227.
Swami, V., Chamorro-Premuzic, T., & Furnham, A. (2010). Unanswered
questions: A preliminary investigation of personality and individual
difference predictors of 9/11 conspiracist beliefs. Applied Cognitive
Psychology,24, 749–761.
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, 443–463.
Swami, V., Pietschnig, J., Tran, U. S., Nader, I. W., Stieger, S., & Voracek,
M. (2013). Lunar lies: The impact of informational framing and individ-
ual differences in shaping conspiracist beliefs about the moon landings.
Applied Cognitive Psychology,27,71–80.
Thorburn, S., & Bogart, L. M. (2005). Conspiracy beliefs about birth
control: Barriers to pregnancy prevention among African Americans of
reproductive age. Health Education & Behavior,32, 474–487.
Van den Bos, K. (2009). Making sense of life: The existential self trying to
deal with personal uncertainty. Psychological Inquiry,20, 197–217.
Van Harreveld, F., Rutjens, B. T., Schneider, I. K., Nohlen, H. U., &
Keskinis, K (2014). In doubt and disorderly: Ambivalence promotes
compensatory perceptions of order. Journal of Experimental Psychology:
General,143, 1666–1676.
Van Prooijen, J.-W., & Jostmann, N. B. (2013). Belief in conspiracy
theories: The influence of uncertainty and perceived morality. European
Journal of Social Psychology,43, 109–115.
Van Prooijen, J.-W., Krouwel, A. P. M., Boiten, M., & Eendebak, L. (2015).
Fear among the extremes: How political ideology predicts negative
emotions and outgroup derogation. Personality and Social Psychology
Bulletin,41, 485–497.
Van Prooijen, J.-W., Krouwel, A. P. M., & Pollet, T. (2015). Political
extremism predicts belief in conspiracy theories. Social Psychological
and Personality Science,6, 570–578.
Van Prooijen, J.-W., & Van Dijk, E. (2014). When consequence size
predicts belief in conspiracy theories: The moderating role of perspective
taking. Journal of Experimental Social Psychology,55,63–73.
Van Prooijen, J.-W., & Van Lange, P. A. M. (2014). The social dimension of
belief in conspiracy theories. In J.-W. van Prooijen, & P. A. M. van Lange
(Eds.), Power, politics, and paranoia: Why people are suspicious of their
leaders (pp. 237–253). Cambridge, UK: Cambridge University Press.
Whitson, J. A., & Galinsky, A. D. (2008). Lacking control increases illusory
pattern perception. Science,322, 115–117.
Wood, M. J., Douglas, K. M., & Sutton, R. M. (2012). Dead and alive:
Beliefs in contradictory conspiracy theories. Social Psychological and
Personality Science,3, 767–773.
Zonis, M., & Joseph, C. M. (1994). Conspiracy thinking in the Middle East.
Political Psychology,15, 443–459.
Control and conspiracy belief 761
Copyright © 2015 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 29: 753–761 (2015)
