Conspiracy Theory Beliefs: Measurement and the Role of Perceived Lack Of

Home , Bermuda Triangle, Cognitive dissonance

Conspiracy theory beliefs: measurement and the role of perceived lack of

control

Ana Stojanov

A thesis submitted for the degree of

Doctor of Philosophy at the University of Otago,

Dunedin, New Zealand

November, 2019 Abstract

Despite conspiracy theory beliefs’ potential to lead to negative outcomes, psychologists have only relatively recently taken a strong interest in their measurement and underlying mechanisms. In this thesis I test a particularly common motivational claim about the origin of conspiracy theory beliefs: that they are driven by threats to personal control. Arguing that previous experimental studies have used inconsistent and potentially confounded measures of conspiracy beliefs, I first developed and validated a new Conspiracy Mentality Scale, and then used it to test the control hypothesis in six systematic and well-powered studies. Little evidence for the hypothesis was found in these studies, or in a subsequent meta-analysis of all experimental evidence on the subject, although the latter indicated that specific measures of conspiracies are more likely to change in response to control manipulations than are generic or abstract measures. Finally, I examine how perceived lack of control relates to conspiracy beliefs in two very different naturalistic settings, both of which are likely to threaten individuals feelings of control: a political crisis over Macedonia’s name change, and series of tornadoes in North America. In the first, I found that participants who had opposed the name change reported stronger conspiracy beliefs than those who has supported it. In the second, participants who had been more seriously affected by the tornadoes reported decreased control, which in turn predicted their conspiracy beliefs, but only for threat-related claims.

Tentatively, I conclude that threats to control can motivate conspiracy ideation, but only under particular conditions, such as when the threat is extreme and a conspiracy theory is available that offers a relevant explanation, although further research is necessary to explore the boundaries of the effects.

Publication note

Chapter 2 has been published in Social Psychology. Stojanov, A., & Halberstadt, J. (2019).

The Conspiracy Mentality Scale: Distinguishing between irrational and rational suspicion. Social Psychology, 50(4 ), 215-232

Chapter 3 is currently under review in British Journal of Social Psychology.

Chapter 4 is currently under review in European Journal of Social Psychology.

Chapter 5 is in preparation for submission to Social Psychological and Personality Science.

Acknowledgements

When I embarked on this journey in 2016 I expected sleepless nights, long days in the office and coming home one day years later, straight after submitting the thesis, not recognizing my children because they have “miraculously” time traveled several years into the future. It was nothing as close to this and the person responsible for this is my primary supervisor professor Jamin Halberstadt, with whom I collaborated impeccably and whom I can’t thank enough. Thank you Jamin – for being available on very short notice, for agreeing to work on a paper with one week deadline, for the wonderful things you wrote in the recommendation letters, for inspiring me to learn new statistic skills, for improving my writing, refining my reasoning, providing resources and advice, pushing me out of my comfort zone, being my role model and - providing me with the opportunity to experience all this at the University of Otago.

Professor Jesse Bering – although somewhat in the background during this journey, knowing that you are on board during this turbulent ride provided the feeling of security and calm. Thank you for the feedback which helped build my confidence as a writer. I am honored to have worked with you.

I would also like to thank the Division of Sciences and the Department of Psychology at the University of Otago that have generously funded my PhD studies and the Open Society

Foundation that has granted me a Civil Society Scholar Award to conduct a study presented in Chapter 5.

A special thanks goes to the Social Cognition Lab RA: Sarah Hill, Sussie Campbel,

Gabriela Visini, Samantha Smith, Ruth J.M. Huges, Masnoon Majeed and Jean Hwee Low.

iii

I am indebted to the members of the Social Cognition Lab for the stimulating discussions and food for thought they provided as well as the peculiar sense of humor that made our Friday lab meetings enjoyable.

Keren – I couldn’t have asked for a better friend and colleague. Thank you for being part of my life and source of support, motivation, inspiration and consolation during the past three years. Vicky and Harry – thank you for the warm welcome and support.

Finally, I would like to thank my family. My mum and dad for setting high expectations for me, believing in me and supporting me in any imaginable way – thank you for always having my best interest in mind – I am lucky to be your daughter. My husband

Zoran for being the best husband in the world – because of you I am a better person. Thank you for embarking on this challenging journey with me and always being by my side. Neda and Ivan – thank you for being the sweet children you are! Thank you for our family time, family hugs and family games that have provided opportunity for incubation of my research ideas. I love you!

Table of Contents

Abstract ...... i Publication note ...... ii Acknowledgements ...... iii Table of Contents ...... v List of Figures ...... vii List of Tables ...... viii Chapter 1: The study of conspiracy theory belief ...... 1 Chapter 2: The conspiracy mentality scale: Distinguishing between irrational and rational suspicion ...... 37 Abstract ...... 39 Study 1: Scale construction ...... 47 Study 2. Predictive validity and confirmatory factor analysis ...... 58 Study 3a. Convergent and divergent validity ...... 65 Study 3b. Test – Retest reliability ...... 72 Study 4. Construct validity across three nations ...... 72 General discussion...... 76 Chapter 3: Does Perceived Lack of Control Lead to Conspiracy Theory Beliefs? ...... 97 Abstract ...... 99 Study 1...... 104 Study 2...... 108 Study 3...... 110 Study 4...... 111 Study 5...... 113 Study 6...... 118 Meta-analyses ...... 121 General Discussion ...... 123 References ...... 127 Appendix ...... 137 Online Supplemental Material ...... 139 Chapter 4: Does lack of control lead to conspiracy beliefs? A meta-analysis ...... 149 Abstract ...... 150 Method ...... 153 Results ...... 160

Discussion ...... 173 Chapter 5: Perceived Lack of Control and Conspiracy Theory Beliefs in the Wake of Political Strife and Natural Disaster ...... 184 Abstract ...... 186 Study 1: North Macedonian Political Crisis, 2018-2019 ...... 190 Study 2: Midwestern American Tornadoes, 2019 ...... 199 Chapter 6: General Discussion...... 219

List of Figures

Chapter 2

Figure 1. Scree plot and parallel analysis results for the original 64 items scale ...... 53

Chapter 3

Figure 1. Forest plot of analysed effect sizes from Studies 1and 3-6 (squares) and the

cumulative effect size (diamond)...... 122

Chapter 4

Figure 1. PRISMA-style flowchart showing selection of studies for meta-analysis on

conspiracy theories as compensatory control literature...... 155

Figure 2. Forest plot of the effect size for the studies in the meta-analysis...... 168

Figure 3. Funnel plot obtained with the trim and fill procedure...... 170

Figure 4. P-curve analysis results ...... 172

Chapter 5

Figure 1. Timeline of events between September 2018 and September 2019 ...... 192

Figure 2. Change in conspiracy theory ideation for the supporters and opposers...... 196

Figure 3. Path analysis from affectedness to conspiracy beliefs via control...... 203

Figure 4. Path analysis from change in perceived control to change in conspiracy beliefs .. 205

vii

List of Tables

Chapter 2

Table 1. The Pattern Matrix of Factor Loadings for the two factors extracted with Principle

Component Analysis ...... 50

Table 2. Item Information Function Values at 15 Values of θ from -2.8 to 2.8 for conspiracy

theory ideation ...... 56

Table 3. Item Information Function Values f at 15 Values of θ from -2.8 to 2.8 for skepticism

...... 57

Table 4. Unstandardized coefficients, standard errors, standardized coefficients and p values

for the conspiracy theories as outcome variables...... 62

Table 5. Unstandardized coefficients, standard errors, standardized coefficients and p values

for the corruption cases as outcome variables...... 63

Table 6. Unstandardized coefficients, standard errors, standardized coefficients and p values

for the examined variables...... 70

Table 7. Results from test of the factorial equivalence of scores from the Conspiracy

Mentality Scale...... 75

Chapter 3

Table 1. Conspiracy theory pairs ...... 115

Table 2. Descriptive statistics by recall task and scenario type...... 121

Table 3. Overview of study design and posterior probabilities for H0 and H1...... 123

Chapter 4

Table 1. Moderator Coding Criteria and number of studies meeting each criteria ...... 159

viii

Table 2. Study description and effect sizes ...... 161

Table 3. Estimate of effect size for each level of each moderators, along with significance test

of moderation effects ...... 166

Chapter 5

Table 1. Means [and 95% CI] for conspiracy theory ideation (CTI) at time 1 (2018) and time

2 (2019)...... 195

Table 2. Matrix of correlations between the conspiracy beliefs, perceived control and

affectedness ...... 202

Table 3. Means and standard deviations and paired samples t-test results ...... 205

Chapter 1: The study of conspiracy theory belief

What is a conspiracy theory?

Emperor Nero caused the great fire of Rome. The witches conspired with the Devil.

The weather is controlled by the government. These three statements refer to events centuries apart, yet they belong to the same class of explanations, known as conspiracy theories.

Despite a long history of such explanations, systematic academic research into people’s belief in conspiracy theories has only begun in the last decade, probably incentivized by the greater visibility of conspiracy theories online and the growing awareness of their negative consequences (Douglas et al., 2019). But what, exactly, is a conspiracy theory? How does it differ from a “real” conspiracy?

Consider the following two statements:

(1) Al Qaeda members conspired to attack the World Trade Center on September 11th

2001 in order to start a holy war against the United States.

(2) The US government conspired to bring forth the attacks on the World Trade

Center so that they could justify the attacks on Afghanistan.

Both explanations about the attacks on 9/11 imply a conspiracy, but most researchers (and lay people), would agree that only the second is a “conspiracy theory.” They would not, however, agree on the reason: although there is consensus that conspiracy theories involve multiple actors working together to achieve a malevolent goal (e.g., Douglas & Leite, 2017; Lantian,

Muller, Nurra & Douglas, 2017; Swami, 2012; van Prooijen & van Dijk, 2014; Wood, 2017;

Wood, Douglas & Sutton, 2012), definitions vary regarding whether conspiracy theories need

1 to be false (e.g., Barron, Morgan, Towell, Altemeyer & Swami, 2014; Pipes, 1996; Swami et al., 2013; Swami, Weis, Lay, Barron & Furnham, 2016), implausible (e.g., Brotherton &

French, 2014; Brotherton & French, 2015; Brotherton & Eser, 2015;), or even whether their epistemological status is relevant (e.g., van Prooijen, in press).

The conceptual ambiguity surrounding conspiracy theories becomes apparent when examining the instruments developed to measure them, which include questions about corruption (e.g., The city council transferred parts of the budget to the bank accounts of others or Members of the city council received money from construction companies to set this plan in motion; van Prooijen and Acker, 2015), paranormal beliefs (e.g., Events in the

Bermuda Triangle constitute evidence of paranormal activity; Graeupner & Coman, 2017), and paranoia (e.g., To what extent do you think your coworker may be connected to you not getting the promotion?; Whitson & Galinsky, 2008, Online Supplementary Materials).

Thus, the first challenge in any rigorous empirical investigation of conspiracy theories is to demarcate the construct and to differentiate it from other, superficially similar ones.

Before I outline my definition of conspiracy theories, I will discuss how conspiracy theories differ from paranormal beliefs, paranoia, and “real” conspiracies.

Conspiracy, paranoia, and paranormal beliefs. Although people who advocate conspiracy theories are often colloquially described as

“paranoid,” conspiratorial thinking and paranoia in fact differ in important ways. Perhaps most notably, conspiracy theories have an intergroup dimension that merely paranoid beliefs lack (Cichocka, Marchlewska, Golec de Zavala & Olechowski, 2015; 2016; Duran, Nicholson

& Dale, 2017; Leone, Giacomantonio & Lauriola, 2017; Mashuri & Zaduqisti, 2015; van

Prooijen & van Lange, 2014.1 The focus of suspicion also differs in the two phenomena: while paranoia is characterized by suspicion towards people in general, conspiracy theories reflect suspicion towards powerful people in particular (Imhoff & Lamberty, 2018). In this sense, they can be considered a form of political attitude (Imhoff & Bruder, 2014).

Moreover, unlike paranoia, conspiracy theories reflect epistemological concerns (Garrett &

Weeks, 2017) and are embedded into a tradition of explanation (Byford, 2011) with a distinct narrative structure, thematic configuration and explanatory logic. Finally, as an empirical matter, paranoia and conspiracy beliefs are relatively independent, with only low to moderate correlations reported in the literature (Brotherton & Eser, 2015; Darwin, Neave & Holmes,

2011; Wilson & Rose, 2013).

Likewise, although there are similarities between conspiracy theories and paranormal beliefs, there are also important differences. For example, while conspiracy beliefs usually arise as a response to a conventional explanation, paranormal beliefs arise because of a lack of conventional explanations (Wood, 2017). Also, while conspiracy theories may be ideologically coloured (Lewandowsky, Gignac & Oberauer, 2013; Uscinski & Parent, 2014), paranormal beliefs generally are not. And, again, the two phenomena are empirically distinguishable: the correlation between them is only moderate (Dagnall et al., 2017; Darwin,

Neave & Holmes, 2011; Drinkwater, Dagnall & Parker, 2012; Lobato, Mendoza, Sims &

Chin, 2014; Swami et al., 2011), and they are associated with different statistical biases

(Dagnall, Denovan, Drinkwater, Parker & Clough, 2017).

1 The difference between individual vs. in-group focal points is most clearly reflected in paranoid schizophrenia, which is a clear case of the former.

Conspiracy theories and plausible allegations of conspiracies It is more difficult to separate conspiracy theories from “real” conspiracies. Although scholars agree that a distinction should be made between them (Bale, 2007; Byford, 2011;

Keeley,1999), there is disagreement about where the line should be drawn. One person’s

“conspiracy theory” may be another person’s reasonable suspicion (Harambam & Aupers,

2017).

For example, when Noam Chomsky was accused of promoting a conspiracy theory that “blames the U.S. government for virtually every ill around the world” (Pipes, 1997), he argued that his analyses clearly differ from conspiracy theories, citing claims about the

Kennedy assassination as an example of the latter. His response aroused the ire of JFK conspiracy theorists, however, who argued that the investigation into who killed President

Kennedy is clearly distinct from “bogus” conspiracy theories (Byford, 2011). Thus, each of the three acknowledged that their topic of inquiry conceptually similar, yet distinctive from conspiracy theories, but failed to agree what the category of conspiracy theories should actually include.

Despite the fluid and politicized nature of conspiracy theory claims, several authors have attempted to differentiate them from what might be called “plausible conspiracies” (e.g. Bale,

2007; Barkun, 2003; Brotherton, 2015; Jane & Flemming, 2014), noting several critical distinguishing features:

Apocalyptic intent. Conspiracy theories typically represent the world in Manichean terms and embed their claims within a broader story about good vs. evil (Bale, 2007; Barkun, 2003;

Brotherton, 2015; Cubitt, 1989; Oliver & Wood, 2014). The actors in conspiracy theories are seen as Evil Incarnate – representatives of a vast, sinister, omnipotent force (Bale, 2007) with far reaching goals (e.g., to “destroy a way of life”; Hofstadter, 1966, p. 29). In contrast, in plausible conspiracies, such as those involving bribery or corruption (Jane & Flemming,

2014), even if the conspiracy itself stems from evil motives, the perpetrators are generally viewed as flawed individuals, with both positive and negative traits, and conflicting, local motives for personal advancement (Bale, 2007; Uscinski & Parent, 2014).

Omnipotence. Actors in conspiracy theories are seen as monolithic, faultless, omnipotent, omnipresent, and farsighted individuals who are in the process of successfully implementing a grand antisocial agenda (Bale, 2007; Pipes, 1996; Sunstein & Vermeule, 2009). Actors in plausible conspiracies, on the other hand, are seen as bound by geographical, temporal and cognitive limitations. They are human beings who may themselves disagree about the intent or appropriateness of their actions (Popper, 2006), and are prone to errors (Harrison &

Thomas, 1997; Sunstein & Vermeule, 2009). Contrast, for example the “deep state” alleged by current American President Donald Trump, with the claim that president Trump himself, with his aides, conspired to extort politically favorable statements from Ukraine. In the former, a highly competent government-within-a-government is claimed to pervade every level of state bureaucracy (omnipresence), with the ultimate goal of overthrowing the legitimately elected government (grand antisocial agenda); in the latter, the actors are flawed, sometimes incompetent, individuals with their own, inconsistent agendas.

Historical driving force. Conspiracy theories are typically concerned with a single all- encompassing plot through which history is explained (Bale, 2007; Byford, 2011; Pipes,

1997), while the roles of chance, coincidences and unintended consequences are downplayed or ignored. In contrast, plausible conspiracies are generally limited in scope and revolve around a local issue (Bale, 2007). Moreover, things rarely unfold according to plan in plausible conspiracies, and thus are likely to fail, and unlikely to affect history in the way conspiracy theorists propose (Brotherton, 2015; Byford, 2011; Cubitt, 1993; Keeley, 1999;

Popper, 2006). Contrast, for example, the supposed role of the Illuminati in the 9/11 attacks, to the role of the Nixon administration in the Watergate scandal, the latter of which is better

5 described as an opportunistic act situated in a particular time and place, than evidence of the workings of a deeper and more sinister force accountable for multiple historical events (Jane

& Flemming, 2014).

Patternicity. Because conspiracy theories tend to represent events as caused by a single, conspiratorial hub, believers can easily find hidden patterns, connections, and meaning between events that, to nonbelievers, seem unrelated (Barkun, 2003; Keeley, 1999; Shermer,

2010). For example, Hurricane Katrina is said to be only a single instance in a much larger plot for weather control (Penre, 2005), which in turn is only one goal of the New World

Order, a clandestine, sinister organization, which has, in addition, staged mass shootings in order to easily pass gun laws (Brotherton, 2015). By contrast, plausible conspiracies are usually seen as isolated events, and believers generally resist the temptation to weave a narrative in which all facts are accounted for.

Unofficial explanations. Conspiracy theories represent attempts to debunk or to identify anomalies in official explanations of events, and are typically contrasted with them

(Aaronovitch, 2009; Brotherton, 2013, 2015; Coady, 2006; Levy, 2007; Keeley, 1999). Thus, the conspiracy theory is always able to account for more data than the official explanation. By contrast, many plausible conspiracies are official explanations of how events unfolded and almost always leave some things unexplained. For example, the official explanation regarding the murder of JFK leaves unexplained why a man was waving an umbrella immediately prior to the president’s assasination. A conspiracy theory however, explains that the “umbrella man” was signalling to Oswald when to fire.

Unfalsifiable claims. Conspiracy theories are essentially irrefutable and unfalsifiable

(Brotherton, 2015; Byford, 2011; Keeley, 1999; Uscinski & Parent, 2014). Lack of evidence for the conspiracy theory is interpreted as support for the theory, because it purportedly demonstrates that the conspirator is concealing their machinations. Similarly, disconfirming

6 evidence is taken as proof of the conspiracy because it is seen as being planted by the conspirators (Buenting & Taylor, 2010). Thus, any argument against the conspiracy theory is turned in favor of it (Keeley, 1999). Indeed, Basham (2003) argues that conspiracy theories are unfalsiable by definition, because gaps in the evidence are exactly what the theory predicts. In contrast, plausible conspiracies can in principle be verified or disproven by providing confirming (or disconfirming) evidence, as was the case, for example, in the Enron scandal, when former CEO Jeffrey Skilling was convicted of conspiracy, fraud, and insider trading.

Although they help to delineate key differences between conspiracy theories and plausible conspiracies, none of the distinctions above offers a foolproof way to assess whether a claim is a conspiracy theory or not. For example, an official story is not necessary for a conspiracy theory to arise. As Brotherton (2015) notes, only several hours after the

Sandy Hook shootings, when no official explanations were yet offered, conspiracy theorists

"knew" that it was a false flag operation by the government intended to gain support for gun control legislation. Complicating efforts to define and identify conspiracy theories is the fact that conspiracy theorists often merge their arguments with conventional political analysis

(Byford, 2011). By citing, for example, gun control legislation as a motive behind the Sandy

Hook conspiracy, or justification for the invasion over Afghanistan and Iraq as the motive behind the 9/11 government conspiracy, conspiracy theorists make their claims more plausible and acceptable (Bost & Prunier, 2013).

It might be argued that the attempt to distinguish conspiracy theories from plausible conspiracies is mostly relevant to so-called world (Pipes, 1997) or systemic (Barkun, 2003) conspiracy theories, which are mainly concerned with global goals, such as control over the world’s banking system, but are less relevant for conspiracy theories concerning single events

(Barkun, 2003), such as the Kennedy assassination or the crash of TWA Flight 800 (Hagen,

2017). However, although these more limited conspiracy theories seem to stand in isolation, they often become drawn into global theories via the dynamics of conspiratorial inquiry.

Postulating a conspiracy theory to explain an event typically raises more questions than it answers (If the conspiracy theory is true, how could a plot of such a scale not have a single leak? Who has the power to pull off such a crime?), and such questions are quickly answered by activists, global conspiracy theorists, and journalists (Byford, 2011), who often recycle ideas from existing theories. Thus, even standalone events, seen by conspiracy theorists through a wide-angle, unfalsifiable explanatory lens, may be perceived as part of a much bigger pattern, and incorporated into an interconnected conspiracy-related worldview.

Working definition of a conspiracy theory Based on the foregoing considerations, I define a conspiracy theory as an implausible explanation of significant social or political events that postulates powerful agents colluding in secret to achieve a malevolent goal, and that contradicts the official explanation made by the relevant epistemic authority.

By restricting conspiracy theories to implausible explanations, I exclude cases such as corruption or bribery, which are plausible allegations of conspiracy; by restricting them to significant social and political events I ensure that the conspiracy is not seen as relevant to the individual alone (a characteristic of paranoia). In addition, limiting the definition to events caused by powerful agents excludes cases that lack clear intentionality (e.g., disappearance of ships in the Bermuda triangle), and requiring malevolent goals omits banal explanations for social events (e.g. board members always have to “agree” on a course of action behind closed doors (i.e. conspire), which may adversely affect its competitors: “Netflix conspired to put

Blockbuster out of business”). Finally, by requiring that conspiracy theories contradict the official claim made by the relevant epistemic authorities (e.g., the U.S. President’s

8 explanation of 9/11), it excludes accepted explanations (e.g., that members of Al Qaeda conspired to cause the air disasters that brought down the Twin Towers and targeted other high-profile U.S. structures).

The definition offered here is both more refined and more conservative than what appears in the literature (e.g. “…explanations that refer to hidden groups working in secret to achieve sinister objectives”; Lantian et al., 2017, p.160), capturing the characteristics that are prototypical of conspiracy theories, while excluding most causal claims that lie outside the phenomena researchers seek to explain.

Why study belief in conspiracy theories

Conspiracy theories appear to be a ubiquitous sociocultural and historical phenomenon (e.g., Baer, 2013; Dieguez, Wagner-Egger & Gauvrit, 2016; Fassin, 2011; Liu,

2015; Mancosu, Vassallo & Vezzoni, 2017; Mezran, 2016; Ortmann & Heathershaw, 2012;

Radnitz, 2015; Swami, 2012). Conspiracy theory beliefs have been present at least since classical antiquity. For example, in the speeches of the Attic orators there are stories about plots against the public interest, the regime, or in international affairs (Roisman, 2006).

Similarly, when the great fire of Rome devastated most of the city in 64 CE, a conspiracy theory spread, blaming emperor Nero for intentionally causing the fire (Maltiz, 2008).

A uniquely modern feature of modern conspiracy theories, however, is that they are spread via the internet and, especially through social media applications, at a pace that is difficult (if not impossible) for the relevant epistemic authorities to curtail and control

(Sharma, Yadav, Yadav & Ferdinand, 2017). Further, because conspiracy theorists usually form online communities, proponents are likely to perceive an overly strong public consensus for their beliefs, and to be exposed to confirming messages that further bolster their own ideas

(Warner & Neville-Shepard, 2014). With popular online social platforms such as Google,

Facebook, and YouTube employing algorithms providing content to consumers similar to what they are already engaged with, their exposure to contrasting claims, and to plausible explanations of the same phenomena, is reduced.

From an applied perspective, the popularity of conspiracy theories, and their tendency to spread online rapidly and unchecked, are not necessarily problematic in themselves.

However, empirical studies have revealed negative social consequences of such beliefs, such as decreased intention to vaccinate one’s (fictitious) child (Jolley & Douglas, 2014a), lowered willingness to limit one’s carbon footprint (Jolley & Douglas, 2014b), reduced intention to engage in politics (Butler, Koopman & Zimbardo, 1995), and decreased trust in the government (Einstein & Glick, 2015; Kim & Cao, 2016). Conspiracy theory endorsement has also been related to discriminatory behaviour (Bilewicz & Krzeminski, 2010; Bilewicz &

Sedek, 2015, Bilewicz, Winiewski, Kofta & Wojcik, 2013). For example, in a study by

Jolley, Meleady, and Douglas (2019) participants who were exposed to pro-conspiracy material about Jewish people (vs. anti-conspiracy materials) reported weaker willingness to vote for a Jewish political candidate, and in another study, conspiracy beliefs mediated the effect of relative deprivation on support for discriminatory policies against Jews, such as restrictions in buying land (Bilewicz & Sedek, 2015).

Increasingly, the same phenomena have been observed outside of the laboratory, with individuals’ acts of violence traced to their uncritical acceptance of conspiracy theory beliefs.

In one notorious recent case, for example, an individual held an employee at a pizza restaurant at gunpoint, falsely believing the establishment to be the centre of paedophilia ring tied to the American Democratic party (“Pizzagate: gunman fires in restaurant at centre of conspiracy”, 2016). At a group level, conspiracy theories may be even more insidious. When believed or endorsed by those in power, such theories can produce flawed public policies, as was the case with South African President Thabo Mbeki, whose belief that HIV does not

10 cause AIDS led him to deny antiretrovirals to pregnant HIV-positive women using public health services (Nattrass, 2007). Even without endorsing them privately, those in power might advocate and manipulate their followers with conspiracy theories (Nefes, 2017), directing the public attention towards imaginary enemies and away from genuine issues of concern (Heins,

2007), a strategy, Gibson (2018) writes, paved the way for Trump’s presidency in 2016.

Believing that the enemy is conspiring for world domination may lead people to adopt

“defensive-aggressive” tactics and goals, exacerbating conflicts (Pipes, 1997) and in extreme cases triggering persecution or even genocide (Behringer, 2004; Cohn, 1967; Fekete, 2011;

Uenal, 2016; Winiewski, Soral & Bilewicz, 2015).

Further, mere exposure to conspiracy theories may increase their endorsement (Banas

& Miller, 2013) and/or undermine the official version of events (Mulligan & Habel, 2013;

Raab, Auer, Ortlieb & Carbon, 2013), even without participants being aware of their effects

(Douglas & Sutton, 2008). Arguably, even the most outlandish, and therefore seemingly innocuous, conspiracy theories, such as that government officials are actually reptilian aliens

(Lewis & Kahn, 2005), are potentially dangerous, in that they might prime participants to think in conspiratorial terms, lower their standards of evidence, or serve as “gateway” conspiracy theories that ultimately connect with the broader, darker conspiracy theory culture, for example by introducing anti-Semitic authors to new generations (Byford, 2011;

Robertson, 2013).

Thus, given the wide appeal of conspiracy theories and their possible negative consequences, it is important to establish the psychological mechanisms that do (or do not) underpin such beliefs. From a practical or applied perspective, this empirical approach may also have currency when considering interventions designed to effectively counteract them.

The Psychology of Conspiracy Theory Beliefs

According to the functionalist perspective (Boden, Barenbaum, & Gross, 2016;

Cummins, 1975, Wright, 1973), people are motivated to hold beliefs that serve particular functions. For example, belief in supernatural agents may stem, in part, from a universal drive to avoid death anxiety (Norenzayan, & Hansen, 2006), or to view the world as a just place

(Lerner, 1980). Similarly, research on conspiracy theory beliefs has pointed to several such functions, linked to a suite of motives which Douglas, Sutton and Cichocka (2017, see also

Douglas, Sutton & Cichocka, 2019) have broadly classified into three categories: epistemic, existential and social.

Epistemic motives In the broadest functionalist terms, a conspiracy theory may stem from the need to understand the world. For example, attribution theory (Heider, 1958) suggests that people search for causal explanations, especially of negative and/or unexpected events (Wong &

Weiner, 1981). A conspiracy theory, with its inherent unfalsifiability (Brotherton, 2015;

Keeley, 1999; Byford, 2011; Uscinski & Parent, 2014), is therefore perfectly suited to satisfy this causal explanatory drive. Consistent with the epistemic account, people who are high in cognitive closure, for whom a desire for a definitive answer, even an implausible one, is preferred to ambiguity (Kruglanski, 1990 a,b), are particularly likely to endorse conspiracy theories when the conspiracy theories are salient and no official explanation is available

(Marchlewska, Cichocka, & Kossowska, 2018). Moreover, a recent study by van Prooijen,

Douglas and de Inocencio (2018) revealed a relationship between adherence to conspiracy beliefs and proneness towards pattern perception, suggesting a tendency for conspiracy theory beliefs among those inclined to “connect the dots,” or to find meaning in noise.

Social motives

Conspiracy theories may also satisfy social motives. For example, there are some indications that conspiracy beliefs are endorsed to fulfil the need to feel unique or to be seen as having access to privileged information (Imhoff & Lamberty, 2017; Lantian et al., 2017).

Conspiracy theories may also be endorsed in an attempt to feel positive about one’s ingroup.

For example, societally marginalized groups such as ethnic minorities (Bogart & Thorburn

2005; Crocker, Luthanen, Broadnax & Blaine, 1999, van Prooijen, Staman & Krouwel, 2018;

Simmons & Parsons, 2005), as well as individuals with low social status (Uscinski & Parent,

2014), endorse conspiracy theories to a higher degree, perhaps as an attempt to narratively frame the ingroup as competent but sabotaged by a hostile outgroup. The need for belongingness has also been related to conspiracy beliefs (Graeupner & Coman, 2017; van

Prooijen 2016). Relatedly, collective narcissists, who are hypervigilant to any indicators of disrespect or insult against the in-group (Grzesiak-Feldman 2015), may endorse conspiracy theories to attenuate their perception that the in-group is under threat from malevolent outside forces seeking to tear their tight-knit community apart (see Golec de Zavala & Federico,

2018).

Existential motives In addition to serving important epistemological and social functions, causal explanations also help people feel safe and secure by providing them with the feeling that things that happen in the social environment are controllable. According to Douglas et al.’s

(2017), conspiracy theories make people feel safer by identifying the conspirators and the threats they pose (Bost & Prunier, 2013). In support of this argument, anxiety has been shown to be positively correlated with conspiracy theory endorsement (see Grzesiak-Feldman,

2013). Conspiracy theory beliefs are also associated with weaker socio-political control

(Bruder, Haffke, Neave, Nouripanah, & Imhoff, 2013), greater anomie (Abalakina-Paap,

Stephan, Craig, & Gregory, 1999; Goertzel, 1994), greater death anxiety (Newheiser, Farias,

& Tausch, 2011) and lower perceived personal control (Sullivan, Landau & Rothschild,

2010). Such findings suggest that conspiracy theories may, at least in part, act as a buffer against existential threats.

Relatedly, compensatory control theory (Laurin, Kay, & Moscovitch, 2008; Kay,

Gaucher, Napier, Callan, & Laurin, 2008), a theoretical framework highlighted in this thesis, holds that when people lack instrumental control, they may believe in conspiracy theories in an attempt to restore the view that the world is ordered and predictable (Kay & Eibach, 2013;

Kay, Sullivan, & Landau, 2015). In the present thesis, I systematically investigate the possibility that conspiracy beliefs can be understood, in part, as an effort at compensatory control.

Compensatory control theory

According to compensatory control theory, having a sense of control satiates a higher order need to perceive the world as structured and ordered (Kay & Sullivan, 2013). If individuals believe that their experiences are the result of their own actions, they are more likely to interpret their environment as ordered and manageable.

Although the need to believe in a nonrandom, orderly world is a chronic motivation

(Heine, Proulx, & Vohs, 2006; Landau, Greenberg, Solomon, Pyszczynski, & Martens,

2006), people do not always exert control, or perceive that they have it. Indeed, in some cases people may prefer not to have control (e.g., Burger & Cooper, 1979; Iyengar & Lepper, 1999;

Ji, Peng & Nisbet, 2000; Wohl & Enzle, 2003). For example, if a person is concerned about their self-presentation, they may avoid leadership roles or be reluctant to take on responsibility for the performance of the group (Burger, 1989). Likewise, those who believe they are not good at a task will not be motivated to see themselves as effective agents who

14 can complete it (Dweck & Leggett, 1988). Furthermore, socioeconomic status correlates with a sense of personal control (Kraus, Piff, & Keltner 2009), and some impoverished communities may be chronically deprived of the feeling of personal agency. Thus, although having a sense of control is ostensibly vital for people’s well-being, people vary in their control beliefs, and regularly experience a lack of control and feelings of helplessness. When the drive for maintaining personal control is frustrated, the world cannot be easily represented as an ordered, predictable place. How do people then manage to satisfy the need for a structured environment if their own causal role in it is unstable?

According to compensatory control theory, although a sense of personal control is the primary mechanism by which our view of the world as an organized, non-random place is maintained, there are other means of achieving this state of explanatory homeostasis.

Specifically, compensatory control theory posits a cognitive substitutability between feelings of internal and external control. That is, when personal agency is left wanting, people nevertheless imbue the world with order and structure by relying on external systems.

Believing that such outside forces are responsible for events beyond the self’s control enables the individual to satisfy their higher-order need for structure and non-randomness, by providing the sense that “someone is in control”.

Much research has confirmed the general hypotheses generated from compensatory control theory. For example, when personal control is experimentally threatened, participants are more likely to believe in a controlling God vs. a “creator” God (Kay et al., 2008; Kay,

Shepherd, Blatz, Chua, & Galinsky, 2010), to believe in precognition (Greenaway, Louis, &

Hornsey, 2013), to perceive patterns in noisy images (Whitson & Galinsky, 2008), to show support for the government (Kay, Gaucher, Napier, Callan, & Laurin, 2008), and to prefer structured scientific theories (e.g., to prefer the idea that moral development is staged rather than a continuum; Rutjens, van Harreveld, van der Pligt, Kreemers, & Noordewier, 2013).

Although this theoretical framework has been applied to conspiracy theory beliefs

(e.g. Kay & Eibach, 2013; Kay, Sullivan & Landau, 2015), it is not clear whether such beliefs should be an effective means of restoring control. For example, as will be argued in Chapter

3, the theory states that people seek to optimize order, not to maximize it, and it could be argued that conspiracy theories provide “too much” order and structure by accounting for errant data, connecting seemingly unconnected events and leaving nothing unexplained

(Clarke, 2002; Keeley, 1999). Further, even if conspiracy theories represent an effective means of restoring order and structure, they would seem, on their face, a means of last resort.

While compensatory control theory does not require that sources of control or meaning be benevolent, people theoretically favour control systems that are both “culturally accessible” and “socially acceptable” (Kay et al., 2008; Kay & Sullivan, 2013). In most—if not all— conspiracy theories, however, the alleged conspirators are malevolent (van Prooijen, 2018), thus making them poor candidates for compensatory processes when numerous effective and socially sanctioned alternatives exist (Kay, Gaucher, Mcgregor & Nash, 2010; Kay, Gaucher,

Napier, Callan, & Laurin, 2008; Kay, Whitson, Gaucher & Galinsky, 2009; Landau, Kay &

Whitson, 2015). Complicating matters further, as argued in Chapter 3, “belief in conspiracy theories,” is not an easily defined construct, and the validity of some attempts to operationalize it is questionable.

The goal of this thesis, therefore, is to revisit the causal relationship between control and conspiracy beliefs, via an analysis of existing evidence, the development and validation of a new measure of conspiracy ideation, and a series of novel laboratory and field studies.

The results have the potential not only to inform and broaden the scope of compensatory control theory, but also to provide insight into the motivational mechanisms of conspiracy theories themselves.

The structure of the thesis and rationale for studies.

Before examining the role of compensatory control in conspiracy beliefs I required an appropriate and validated measure of the latter. Currently, researchers usually measure conspiracy beliefs in one of two general ways: as endorsement of generic, content-free beliefs about the world; or as endorsement of specific conspiracy theories. However, as I argue in the next chapter, existing instruments have a number of shortcomings, including content validity issues and dubious factor structures. Thus, as a first step in the research programme, I sought to develop in Chapter 2 a validated measure of generic conspiracy beliefs: the Conspiracy

Mentality Scale (CMS; Stojanov & Halberstadt, 2019). Next, in Chapter 3, I present a series of studies to test the hypothesis that lack of control increases conspiracy beliefs as measured on the CMS, as well as on other measures of conspiracy beliefs to establish the robustness of the findings.

To foreshadow, the studies in Chapter 3 revealed little evidence for the compensatory control hypothesis. Consequently, I return in Chapter 4 to the literature to provide a systematic meta-analysis of all experimental manipulations, published or unpublished, of control, in order to assess the robustness of my findings in the context of the entire literature.

The results from the meta-analysis mainly supported the conclusion in Chapter 2: there is little evidence for the control – conspiracy beliefs link, at least as these constructs are manipulated and measured in the laboratory.

In Chapter 5, I consider the possibility that control threats are inherently difficult to invoke (or observe) under laboratory conditions. Therefore, in this chapter, I present a novel pair of naturalistic studies in which I compared participants’ conspiracy beliefs during periods of political crisis or natural disaster, when sense of control should be low (Afifi, Afifi, &

Merrill, 2014), to periods when their control was not threatened (or had been restored). The

17 results provided some evidence that the control-threatening event caused belief in at least a subset of conspiracy beliefs.

Finally, in Chapter 6, I integrate the findings from Chapter 2-5 and tentatively conclude that perceived lack of control may indeed affect conspiracy beliefs, but only under certain conditions, such as when one experiences intense threat in the presence conspiracy claims that relate directly to it. This domain specificity is discussed in terms of its implications for compensatory control theory, and for future research opportunities. Finally, I speculate about the possibilities for reducing conspiracy beliefs, such as by promoting faith in existing institutions or by empowering the self.

References: Afifi, W. A., Afifi, T. D. & Merrill, A. (2014). Uncertainty and Control in the Context of a

Category‐Five Tornado. Research in Nursing and Health, 37, 358-366.

oi:10.1002/nur.21613

Aaronovitch, D. (2009). Voodoo histories: The role of the conspiracy theory in shaping

modern history. London: Jonathan Cape.

Abalakina-Paap, M., Stephan, W.G., Craig, T., & Gregory, W.L. (1999). Beliefs in

conspiracies. Political Psychology, 20 (3), 637-647.

Baer, M.D. (2013). An enemy old and new: The Dönme, anti-Semitism, and conspiracy

theories in the Ottoman Empire and Turkish Republic. Jewish Quarterly Review, 103

(4), 523-555.

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.

Banas, J. A., & Miller, G. (2013). Inducing resistance to conspiracy theory propaganda:

Testing inoculation and metainoculation strategies. Human Communication

Research, 39(2), 184-207. doi:10.1111/hcre.12000

Barkun, M. (2003). A Culture of Conspiracy: Apocalyptic Visions in Contemporary America.

Berkeley, CA: University of California Press.

Barron, D., Morgan, K., Towell, T., Altemeyer, B., & Swami, V. (2014). Associations

between schizotypy and belief in conspiracist ideation. Personality and Individual

Differences, 70, 156-159. doi:10.1016/j.paid.2014.06.040

Basham, L. (2003). Malevolent global conspiracy. Journal of Social Philosophy, 34, 91-103.

doi:10.1111/1467-9833.00167

Behringer, W. (2004). Detecting the ultimate conspiracy, or how Waldensians became

witches, in conspiracies and conspiracy theories in early modern Europe (Coward,

Barry and Swann Julian, Eds.). Hampshire, England: Ashgate.

Bilewicz, M., & Krzeminski, I. (2013). Anti-Semitism in Poland and Ukraine: the belief in

Jewish control as a mechanism of scapegoating. International Journal of Conflict and

Violence, 4, 234-243.

Bilewicz, M., & Sedek, G. (2015). Conspiracy stereotypes: Their sociopsychological

antecedents and consequences. In M. Bilewicz, A. Cichocka, & W. Soral (Eds.), The

psychology of conspiracy (pp. 3-22). New York, NY, US: Routledge/Taylor & Francis

Group.

Bilewicz, M., Winiewski, M., Kofta, M. & Wojcik, A. (2013). Harmful ideas, The structure

and consequences of anti–Semitic beliefs in Poland. Political Psychology, 34, 821-

839.

Boden, M. T., Berenbaum, H., & Gross, J. J. (2016). Why do people believe what they do? A

functionalist perspective. Review of General Psychology, 20(4), 399–

411. https://doi.org/10.1037/gpr0000085

Bogart, L. M., & Thorburn, S. (2005). Are HIV/AIDS conspiracy beliefs a barrier to HIV

prevention among African Americans? JAIDS Journal of Acquired Immune

Deficiency Syndromes, 38(2), 213-218. http://dx.doi.org/10.1097/00126334-

200502010-00014

Bost, P. R., & Prunier, S. G. (2013). Rationality in conspiracy beliefs: The role of perceived

motive. Psychological Reports, 113(1), 118-128.

http://dx.doi.org/10.2466/17.04.PR0.113x17z0

Brotherton, R. ( 2015). Suspicious minds: Why we believe conspiracy theories. London:

Bloomsbury.

Brotherton, R. (2013). Towards a definition of “conspiracy theory”. PsyPAG Quarterly, 88

(3), 9–14.

Brotherton, R., & Eser, S. (2015). Bored to fears: Boredom proneness, paranoia, and

conspiracy theories. Personality and Individual Differences, 80, 1–5.

Brotherton, R., & French, C. C. (2014). Belief in conspiracy theories and susceptibility to the

conjunction fallacy. Applied Cognitive Psychology, 28(2), 238-248.

doi:10.1002/acp.2995

Brotherton, R.J., & French, C.C. (2015). Intention seekers: conspiracist ideation and biased

attributions of intentionality. PloS one.

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, Article ID 225.

http://dx.doi.org/10.3389/fpsyg.2013.00225

Buening, J. & Taylor, J. (2010). Conspiracy theories and fortuitous data. Philosophy of the

Social Sciences, 40, 567-578.

Burger, J. M. (1989). Negative reactions to increases in perceived personal control. Journal of

Personality and Social Psychology, 56(2), 246-256.http://dx.doi.org/10.1037/0022-

3514.56.2.246

Burger, J. M., & Cooper, H. M. (1979). The desirability of control. Motivation and Emotion,

3(4), 381-393.http://dx.doi.org/10.1007/BF00994052

Butler, L.D., Koopman, C., & Zimbardo, P.G. (1995). The psychological impact of viewing

the film “JFK”: Emotions, beliefs, and political behavioral intentions. Political

Psychology, 16(2), 237-257.

Byford, J. (2011). Conspiracy theories: A critical introduction. Basingstoke UK: Palgrave

MacMillan.

Cichocka, A., Marchlewska, M., Golec de Zavala, A., & Olechowski, M. (2016). ‘They will

not control us’: Ingroup positivity and belief in intergroup conspiracies. British

Journal of Psychology, 107, 556–576.

Cichocka, A., Marchlewska, M., Golec de Zavala, A., & Olechowski, M. (2015). Grandiose

delusions. Collective narcissism, secure in-group identification, and belief in

conspiracies. In M. Bilewicz, A. Cicochka & W. Soral (Eds.), The Psychology of

Conspiracy Theories. London: Routledge.

Clarke, S. (2002). Conspiracy Theories and Conspiracy Theorizing. Philosophy of the Social

Sciences, 32(2), 131–150. https://doi.org/10.1177/004931032002001

Coady, D. (2006). An introduction to the philosophical debate about conspiracy theories. In

D. Coady (Ed.), The philosophy of conspiracy theories. Burlington, VT: Ashgate.

Cohn, N. (1967). Warrant for genocide. London: Eyre & Spottiswoode.

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(8), 941-953.

http://dx.doi.org/10.1177/01461672992511003

Cubitt, G. (1989). Conspiracy myths and conspiracy theories. Journal of the Anthropological

Society of Oxford, 20(1), 12-26.

Cubitt, G. (1993). The Jesuit myth: Conspiracy theory and politics in nineteenth-century

France. Oxford, UK: Clarendon Press.

Cummins, R. (1975). Functional analysis. The Journal of Philosophy, 72, 741–764.

10.2307/2024640

Dagnall, N., Denovan, A., Drinkwater, K., Parker, A., & Clough, P. (2017). Statistical bias

and endorsement of conspiracy theories. Applied Cognitive Psychology, 31, 368-378.

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(8), 1289-1293. doi:10.1016/j.paid.2011.02.027

Dieguez, S., Wagner-Egger, P., & Gauvrit, N. (2016). Nothing happens by accident, or does

it? A low prior for randomness does not explain belief in conspiracy theories.

Psychological Science, 26(11), 1762-1770.

Douglas, K. M., & Sutton, R. M. (2008). The hidden impact of conspiracy theories:

Perceived and actual influence of theories surrounding the death of Princess Diana.

Journal of Social Psychology, 148, 210–221. doi: 10.3200/SOCP.148.2.210-222

Douglas, K. M., & Leite, A. C. (2017). Suspicion in the workplace: Organizational

conspiracy theories and work-related outcomes. British Journal of

Psychology, 108(3), 486-506. doi:10.1111/bjop.12212

Douglas, K. M., Sutton, R. M., & Cichocka, A. (2017). The psychology of conspiracy

theories. Current Directions in Psychological Science, 26(6), 538-542.

doi:10.1177/0963721417718261

Douglas, K. M., Sutton, R. M., & Cichocka, A. (2019). Belief in conspiracy theories: Looking

beyond gullibility. In J. Forgas, & R. Baumeister, (Eds.), The Social Psychology of

Gullibility: Conspiracy Theories, Fake News and Irrational Beliefs (pp. 61-76). New

York, NY: Routledge. doi:10.4324/9780429203787

Douglas, K. M., Uscinski, J. E., Sutton, R. M., Cichocka, A., Nefes, T., Ang, C. S., &

Deravi,

F. (2019). Understanding conspiracy theories. Political Psychology, 40(S1), 3-35. doi:10.1111/pops.12568

Drinkwater, K., Dagnall, N., Parker, A. (2012). Reality testing, conspiracy theory, and

paranormal beliefs. The Journal of Parapsychology, 76(1), 57-77.

Duran, N. D., Nicholson, S. P., & Dale, R. (2017). The hidden appeal and aversion to political

conspiracies as revealed in the response dynamics of partisans. Journal of

Experimental Social Psychology, 73, 268-

278.http://dx.doi.org/10.1016/j.jesp.2017.07.008

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and

personality. Psychological Review, 95(2), 256-273. http://dx.doi.org/10.1037/0033-

295X.95.2.256

Einstein, K. L., & Glick, D. M. (2015). Do I think BLS data are BS? The consequences of

conspiracy theories. Political Behavior, 37(3), 679-

701.http://dx.doi.org/10.1007/s11109-014-9287-z

Fassin, D. (2011). The politics of conspiracy theories: On AIDS in South Africa and a few

other global plots. The Brown Journal of World Affairs, 17(2), 39-50. Retrieved from

http://www.jstor.org/stable/24590791

Fekete, L. (2011). The Muslim Conspiracy Theory and the Oslo Massacre. Race and Class,

53, 30-47.

Garrett, R.K., & Weeks, B.E. (2017). Epistemic beliefs’ role in promoting misperceptions and

conspiracist ideation. PLoS ONE 12(9): e0184733.

ttps://doi.org/10.1371/journal.pone.0184733

Gibson, T. A. (2018). The post-truth double helix: Reflexivity and mistrust in local politics.

International Journal of Communication, 12, 3167-3185

Goertzel, T. (1994). Belief in conspiracy theories. Political Psychology, 15(4), 731-742.

doi:10.2307/3791630

Golec de Zavala, A. & Federico, C. M. (2018), Collective narcissism and the growth of

conspiracy thinking over the course of the 2016 United States presidential election: A

longitudinal analysis. European Journal of Social Psychology, 48, 1011-1018.

doi:10.1002/ejsp.2496

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. doi:10.1016/j.jesp.2016.10.003

Greenaway, K.H., Louis, W.R., & Hornsey, M.J. (2013). Loss of control increases belief in

precognition and belief in precognition increases control. PLoS ONE 8(8): e71327.

https://doi.org/10.1371/journal.pone.0071327

Grzesiak-Feldman, M. (2013). The effect of high-anxiety situations on conspiracy

thinking. Current Psychology, 32(1), 100-118. doi:10.1007/s12144-013-9165-6

Grzesiak-Feldman, M. (2015). Are the high authoritarians more prone to adopt conspiracy

theories? The role of right wing authoritarianism in conspiratorial thinking. In M.

Bilewicz, A. Cichocka and W. Soral (Eds.), The psychology of conspiracy. London:

Routledge.

Hagen, K. (2017): Conspiracy theories and the paranoid style: Do conspiracy theories posit

implausibly vast and evil conspiracies? Social Epistemology, 32 (1), 24-40. DOI:

10.1080/02691728.2017.1352625

Harambam, J. & Aupers, S. (2017). ‘I am not a conspiracy theorist’: Relational

identifications in the Dutch conspiracy milieu. Cultural Sociology, 11(1),113-129.

Harrison, A. A., & Thomas, J. M. (1997). The Kennedy assassination, unidentified flying

objects, and other conspiracies: Psychological and organizational factors in the

perception of “cover-up.” Systems Research in the Behavioral Sciences, 14, 113-128.

Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.

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.

Heins, V. (2007). Critical theory and the traps of conspiracy thinking. Philosophy & Social

Criticism, 33(7), 787–801. https://doi.org/10.1177/0191453707081675

Hofstadter, R. (1966). The paranoid style in American politics. In R. Hofstadter (Ed.), The

paranoid style in American politics and other essays (pp. 3 – 40). New York, NY:

Knopf.

Imhoff, R., & Bruder, M. (2014). Speaking (un-)truth to power: Conspiracy mentality as a

generalised political attitude. European Journal of Personality, 28(1), 25-43.

doi:10.1002/per.1930

Imhoff, R., & Lamberty, P. (2018). How paranoid are conspiracy believers? toward a more

fine-grained understanding of the connect and disconnect between paranoia and belief

in conspiracy theories. European Journal of Social Psychology, 48(7), 909-926.

doi:10.1002/ejsp.2494

Imhoff, R., & Lamberty, P. K. (2017). Too special to be duped: Need for uniqueness

motivates conspiracy beliefs. European Journal of Social Psychology, 47(6), 724-734.

doi:10.1002/ejsp.2265

Iyengar, S. S., & Lepper, M. R. (1999). Rethinking the value of choice: A cultural perspective

on intrinsic motivation. Journal of Personality and Social Psychology, 76(3), 349-

366. http://dx.doi.org/10.1037/0022-3514.76.3.349

Jane, E., & Fleming, C. (2014). Modern conspiracy: The importance of being paranoid. New

York: Bloomsbury Publishing.

Ji, L.-J., Peng, K., & Nisbett, R. E. (2000). Culture, control, and perception of relationships in

the environment. Journal of Personality and Social Psychology, 78(5), 943-

955.http://dx.doi.org/10.1037/0022-3514.78.5.943

Jolley, D. , Meleady, R. and Douglas, K. M. (2019), Exposure to intergroup conspiracy

theories promotes prejudice which spreads across groups. British Journal of

Psychology. Advance online article. doi:10.1111/bjop.12385

Jolley, D., & Douglas, K.M. (2014a). The effects of anti-vaccine conspiracy theories on

vaccination intentions. PLOS ONE, 9(2): e89177. doi:

10.1371/journal.pone.0089177;

Jolley, D., & Douglas, K.M. (2014b). The social consequences of conspiracism: Exposure to

conspiracy theories decreases the intention to engage in politics and to reduce one’s

carbon footprint. British Journal of Psychology, 105, 35-56. doi: 10.1111/bjop.12018

Kay, A. C., & Eibach, R. P. (2013). Compensatory control and its implications for ideological

extremism. Journal of Social Issues, 69(3), 564-585. doi:10.1111/josi.12029

Kay, A. C., Gaucher, D., McGregor, I., & Nash, K. (2010). Religious Belief as Compensatory

Control. Personality and Social Psychology Review, 14(1), 37–48.

https://doi.org/10.1177/1088868309353750

Kay, A. C., Gaucher, D., Napier, J. L., Callan, M. J., & Laurin, K. (2008). God and the

government: Testing a compensatory control mechanism for the support of external

systems. Journal of Personality and Social Psychology, 95(1), 18-35.

http://dx.doi.org/10.1037/0022-3514.95.1.18

Kay, A. C., Shepherd, S., Blatz, C. W., Chua, S. N., & Galinsky, A. D. (2010). For God (or)

country: The hydraulic relation between government instability and belief in religious

sources of control. Journal of Personality and Social Psychology, 99(5), 725–739.

Kay, A.C. & Sullivan, D. (2013). Cultural unity and diversity in compensatory control

processes. In M.J. Gelfand, C. Chiu, & Y. Hong (Eds.), Advances in culture and

psychology, Vol. 3 (pp. 181-226). New York: Oxford University Press.

Kay, A. C., Sullivan, D., & Landau, M. J. (2015). Psychological importance of beliefs in

control and order: Historical and contemporary perspectives in social and personality

psychology. In M. Mikulincer, P. R. Shaver, E. Borgida, & J. A. Bargh (Eds.), APA

handbooks in psychology. APA handbook of personality and social psychology, Vol. 1.

Attitudes and social cognition (pp. 309-337). Washington, DC, US: American

Psychological Association. http://dx.doi.org/10.1037/14341-010

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. https://doi.org/10.1111/j.1467-

8721.2009.01649.x

Keeley, B. (1999). Of conspiracy theories. Journal of Philosophy 96, 109–26.

doi: 10.2307/2564659

Kim, M. & Cao, X. (2016). The impact of exposure to media messages promoting

government conspiracy theories on distrust in the government: Evidence from a two-

stage randomized experiment. International Journal of Communication, 10, 3808-

3827.

Kraus, M.W., Piff, P.K., & Keltner, D. (2009). Social class, sense of control, and social

explanation. Journal of Personality and Social Psychology, 97, 992-1004 .

Kruglanski, A. (1990). Lay epistemic theory in social-cognitive psychology. Psychological

Inquiry, 1(3), 181-197.

Kruglanski, A. W. ( 1990). Motivations for judging and knowing: Implications for causal

attribution. In E. T. Higgins, & R. M. Sorrentino (Eds.), The handbook of motivation

and cognition: Foundation of social behavior Vol. 2 (pp. 333– 368). New York:

Guilford.

Landau, M. J., Greenberg, J., Solomon, S., Pyszczynski, T., & Martens, A. (2006). Windows

into nothingness: Terror management, meaninglessness, and negative reactions to

modern art. Journal of Personality and Social Psychology, 90(6), 879-892.

http://dx.doi.org/10.1037/0022-3514.90.6.879

Landau M. J., Kay A. C., Whitson J. A. (2015). Compensatory control and the appeal of a

structured world. Psychological Bulletin, 141, 694–722.

Lantian, A., Muller, D., Nurra, C., & Douglas, K. M. (2017). "I know things they don't

know!" the role of need for uniqueness in belief in conspiracy theories. Social

Psychology, 48(3), 160-173. doi:10.1027/1864-9335/a000306

Laurin, K., Kay, A. C., & Moscovitch, D. A. (2008). On the belief in God : Towards an

understanding of the emotional substrates of compensatory control. Journal of

Experimental Social Psychology, 44, 1559–1562.

https://doi.org/10.1016/j.jesp.2008.07.007

Leone, L., Giacomantonio, M., & Lauriola, M. (2019). Moral foundations, worldviews, moral

absolutism and belief in conspiracy theories. International Journal of

Psychology, 54(2), 197-204. doi:10.1002/ijop.12459

Lerner. M. (1980). The belief in a just world: a fundamental delusion. New York: Plenum

Press.

Levy, N. (2007). Radically socialized knowledge and conspiracy theories. Episteme, 4(2),

181-192.

Lewandowsky, S., Gignac, G.E., & Oberauer, K. (2013). The role of conspiracist ideation and

worldviews in predicting rejection of science. PLoS ONE 8(10): e75637.

doi:10.1371/journal.pone.0075637

Lewis, T., & Kahn, R. (2005). The reptoid hypothesis: Utopian and dystopian

representational motifs in David Icke’s alien conspiracy theory. Utopian Studies,

16, 45–74.

Liu, J. C.-E. (2015). Low carbon plot: Climate change skepticism with Chinese

characteristics. Environmental Sociology, 1(4), 280–292.

Lobato, E., Mendoza, J., Sims, V., & Chin, M. (2014). Examining the relationship between

conspiracy theories, paranormal beliefs, and pseudoscience acceptance among a

university population. Applied Cognitive Psychology, 28, 617-625.

Maltiz, J. (2008). Nero. Oxford, UK: Blackwell Publishing.

Mancosu, M., Vassallo, S., & Vezzoni, C. (2017). Believing in conspiracy theories: Evidence

from an exploratory analysis of Italian survey data. South European Society and

Politics, 22(3), 327-344.

Marchlewska, M., Cichocka, A., and Kossowska, M. ( 2018) Addicted to answers: Need for

cognitive closure and the endorsement of conspiracy beliefs. European Journal of

Social Psychology, 48, 109– 117. https://doi.org/10.1002/ejsp.2308.

Mashuri, A. & Zaduqisti, E. (2015). The effect of intergroup threat and social identity

salience on the belief in conspiracy theories over terrorism in Indonesia: Collective

angst as mediator. International Journal of Psychological Research, 8, 24-35.

Mezran, K. (2016). Conspiracism in and around Libya. The International Spectator, 51, 113-

118, DOI: 10.1080/03932729.2016.1160685

Mulligan, K. & Habel, P. (2013). The implications of fictional media for political beliefs.

American Politics Research, 41(1), 122-146.

Nattrass, N. (2007). AIDS Denialism vs. Science. Skeptical Inquirer, 31.5. Retrieved on Feb

10th 2018 from https://www.csicop.org/si/show/aids_denialism_vs._science

Nefes, T.S. (2017). The impact of the Turkish government’s conspiratorial framing of the

Gezi Park protests. Social Movement Studies, 16, 610-622.

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(8), 1007–1011.

Norenzayan, A. , & Hansen, I. G. (2006). Belief in supernatural agents in the face of death.

Personality and Social Psychology Bulletin, 32, 174–187.

10.1177/0146167205280251

Oliver, E.J. & Wood, T.J. (2014). Conspiracy theories and the paranoid style(s) of mass

opinion. American Journal of Political Science, 58(4), 952-966.

Ortman, S. & Heathershaw, J. (2012). Conspiracy theories in the post-Soviet space. Russian

Review, 71 (4), 551-564.

Penre, W. (2005, Sep 6). Hurricane Katrina: One Thing to Remember! Retrieved from

http://www.illuminati-news.com/hurricane-katrina.htm

Pipes, D. (1996). The hidden hand: Middle East fears of conspiracy. New York: St. Martin's

Press.

Pipes, D. (1997). Conspiracy: How the paranoid style flourishes and where it comes from.

New York, NY: Simon & Schusters.

Pizzagate: gunman fires in restaurant at centre of conspiracy. (2016). Retrieved September

29, 2019, from BBC website: https://www.bbc.com/news/world-us-canada-38205885.

Popper, K. ( 2006). The conspiracy theory of society. In D. Coady (Ed.), The Philosophy of

Conspiracy Theories (pp. 13-15). Burlington, VT : Ashgate. van Prooijen, J. ‐W. ( 2016) Sometimes inclusion breeds suspicion: Self-uncertainty and

belongingness predict belief in conspiracy theories. Europeam Joural of Social

Psychology, 46, 267– 279. doi: 10.1002/ejsp.2157.

Raab, M. H., Auer, N., Ortlieb, S. A., & Carbon, C.-C. (2013). The Sarrazin effect: the

presence of absurd statements in conspiracy theories makes canonical information less

plausible. Frontiers in Psychology, 4, 453. http://doi.org/10.3389/fpsyg.2013.00453

Radnitz, S. (2015). Paranoia with a purpose: conspiracy theory and political coalitions in

Kyrgyzstan. Post–Soviet Affairs, 32 (5), 474-489.

Robertson, D.G (2013). David Icke’s reptilian thesis and the development of new age

theodicy. International Journal for the Study of New Religions 4, 27–47.

Roisman, J. (2006). The rhetoric of conspiracy in ancient Athens. Berkeley: University of

California Press.

Rutjens, B. T., van Harreveld,, van der Pligt, J., Kreemers, L.M., & Noordewier, M.K.

(2013). Steps, Stages, and Structure : Finding Compensatory Order in Scientific

Theories. Journal of Experimental Social Psychology, 142, 313–318.

https://doi.org/10.1037/a0028716

Sharma, M., Yadav, K., Yadav, N., & Ferdinand, K. C. (2017). Zika virus pandemic –

analysis of Facebook as a social media health information platform. American Journal

of Infection Control, 45(3), 301–302.

Shermer, M. (2010, December). The conspiracy theory detector. Scientific American, 303(6),

102.

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. doi:10.1111/j.0038-4941.2005.00319.x

Stojanov, A., & Halberstadt, J. (2019). The conspiracy mentality scale: Distinguishing

between irrational and rational suspicion. Social Psychology, 50(4), 215-232.

doi:10.1027/1864-9335/a000381

Sullivan, D., Landau, M. J., & Rothschild, Z. K. (2010). An existential function 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(3), 434-449. http://dx.doi.org/10.1037/a0017457

Sunstein, C. R. & Vermeule, A. (2009). Conspiracy theories: Causes and cures. Journal of

Political Philosophy, 17, 202–227. doi: 10.1111/j.1467-9760.2008.00325.x

Swami, V. (2012). Social psychological origins of conspiracy theories: The case of the Jewish

conspiracy theory in Malaysia. Frontiers in Psychology, 3, 280.

http://doi.org/10.3389/fpsyg.2012.00280

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.

doi:10.1111/j.2044-8295.2010.02004.x

Swami, V., Pietschnig, J., Tran, U. S., Nader, I., Stieger, S. & Voracek, M. (2013). Lunar

Lies: the impact of informational framing and individual differences in shaping

conspiracist beliefs about the moon landings. Applied Cognitive Psychology, 27, 71–

80. DOI: https://doi.org/10.1002/acp.2873

Swami, V., Weis, L., Lay, A., Barron, D., & Furnham, A. (2016). Associations between belief

in conspiracy theories and the maladaptive personality traits of the personality

inventory for DSM-5. Psychiatry Research, 236, 86-90.

doi:10.1016/j.psychres.2015.12.027

Uenal, F. (2016). The “Secret Islamization” of Europe: Exploring integrated threat theory for

predicting Islamophobic conspiracy stereotypes. International Journal of Conflict and

Violence, 10(1), 94-108.

Uscinski, J. E., & Parent, J. M. (2014). American conspiracy theories. New York: Oxford

University Press.

Van Prooijen, J.-W. (2018). The psychology of conspiracy theories. Oxon, UK: Routledge. van Prooijen, J.-W., & Acker, M. (2015). The influence of control on belief in conspiracy

theories: Conceptual and applied extensions. Applied Cognitive Psychology, 29, 753-

761.

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. doi:10.1016/j.jesp.2014.06.006 van Prooijen, J.-W., Douglas, K. M., & De Inocencio, C. (2018). Connecting the dots:

Illusory pattern perception predicts belief in conspiracies and the

supernatural. European Journal of Social Psychology, 48(3), 320-335.

doi:10.1002/ejsp.2331 van Prooijen, J.-W., Staman, J., & Krouwel, A. P. M. (2018). Increased conspiracy beliefs

among ethnic and muslim minorities. Applied Cognitive Psychology, 32(5), 661-667.

doi:10.1002/acp.3442

Van Prooijen, J.-W. (in press). An existential thread model of conspiracy theories. European

Psychologist.

34 van Prooijen, J.-W., & van Lange, P. A. M. (2014). The social dimension of belief in

conspiracy theories. In J.-W. van Prooijen and 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.

Warner, B.R. & Neville-Shepard, R. (2014). Echoes of a conspiracy: Birthers, truthers, and

the cultivation of extremism. Communication Quarterly, 62(1), 1-17.

Whitson, J.A., & Galinsky, A.D. (2008). Lacking control increases illusory pattern

perception. Science, 322 5898, 115-7 .

Wilson, M. W. & Rose, C. (2014). The role of paranoia in a dual-process motivational model

of conspiracy belief. In van Prooijen & van Lange (Eds.), Power Politics and

Paranoia: Why People are Suspicious of Their Leaders (273-291). Cambridge:

Cambridge University Press.

Winiewski, M., Soral, W., & Bilewicz, M. (2015). Conspiracy theories on the map of

stereotype content: Survey and historical evidence. In M. Bilewicz, A. Cichocka, &

W. Soral (Eds.), The psychology of conspiracy (pp. 23-41). New York, NY, US:

Routledge/Taylor & Francis Group.

Wohl, M. J. A., & Enzle, M. E. (2003). The effects of near wins and near losses on self-

perceived personal luck and subsequent gambling behavior. Journal of Experimental

Social Psychology, 39(2), 184-191. http://dx.doi.org/10.1016/S0022-1031(02)00525-5

Wong, P.T. & Weiner, B. (1981). Why people ask “why” questions and the heuristics of

attributional search. Journal of Personality and Social Psychology, 40, 650-663.

Wood, M. J. (2017). Conspiracy suspicions as a proxy for beliefs in conspiracy theories:

Implications for theory and measurement. British Journal of Psychology, 108(3), 507-

527. doi:10.1111/bjop.12231

Wood, M. J., Douglas, K. M., & Sutton, R. M. (2012). Dead and alive: Beliefs in

contradictory conspiracy theories. Social Psychological and Personality Science, 3(6),

767-773. doi:10.1177/1948550611434786

Wright, L. (1973). Functions. The Philosophical Review, 82, 139–168. 10.2307/2183766

Chapter 2: The conspiracy mentality scale: Distinguishing

between irrational and rational suspicion

The conspiracy mentality scale:

Distinguishing Between Irrational and Rational Suspicion

Ana Stojanov,1 Jamin Halberstadt 1

1Department of Psychology, University of Otago, Dunedin, New Zealand

Abstract

Belief in conspiracy theories, generally considered to be a unidimensional construct, is associated with negative outcomes. The existing measures of conspiracy theory beliefs have number of shortcomings. We present the development of a novel measure of the tendency to believe in conspiracy theories and report the discovery of a second factor that reflects rational skepticism. In Study 1(N=500) we use item response theory to devise the final items. In Study

2 (N=202) we demonstrate the predictive validity of the two factors for different types of conspiracies. In Study 3 (N=308) we demonstrate convergent/divergent validity. In study 4

(N= 800) we demonstrate construct validity in three countries. Implications for the concept of conspiracy theory and conspiracy theory beliefs are discussed.

Keywords: scale constructions, conspiracy theory beliefs, item response theory, suspiciousness, rational skepticism, conspiracy mentality, construct validity

A powerful secret elite with a sinister agenda is ruling the world. Billionaires have the presidents of the United States on their payroll so that they do exactly what they want them to. Global warming is made up and is a way to channel money to scientists. Such claims of powerful groups working together in secret, which a significant minority of people believe, are prototypical examples of “conspiracy theories.”

Studying conspiracy theory beliefs is important, among other reasons, because people act on them, sometimes with tragic consequences. For example, people who were exposed to the conspiracy theory that “Big Pharma” produces harmful vaccines were less likely to express intention to vaccinate their children (Jolley & Douglas, 2014a), and those who were made to believe that global warming is a hoax were less willing to reduce their carbon footprint (Jolley & Douglas, 2014b). Similarly, belief in the secret Islamization of Europe has been cited as partial motivation for Breivik’s massacre in Norway (e.g. Fekete, 2011; Uenal,

2016), and the belief in authorities conspiring to remove federal rights has been cited as a reason for the Oklahoma bombing (e.g. Knight, 2000).

Studying conspiracy theory beliefs is also important from a theoretical perspective. In particular, given the wide appeal of conspiracy theories (e.g. Jensen, 2013, 2016) it is important to examine what psychological function they may serve. Although there is not yet consensus on what those functions are – some studies suggest that fulfilling the need for uniqueness (Imhoff & Lamberty, 2017; Lantian, Muller, Nurra & Douglas, 2017) or coping with existential threat (Newheiser, Farias & Tausch, 2011) are important factors – it is clear that understanding why people gravitate to conspiracy theories is a key to reducing their harm, and to a broader understanding human motivation.

However, not all conspiracy claims are conspiracy theories, and not all have the same potential for societal harm, or for psychological insight. Indeed, many conspiracy claims, such as allegations of bribery among public officials, are mundane and widely accepted as

40 plausible (and illegal). Even some extreme claims, such as the Watergate conspiracy, are true, and their acceptance unremarkable. Thus, some operational definitions of “conspiracy theory” used in the literature, such as Dentith’s (2014) – “any explanation of an event that cites the existence of a conspiracy as a salient cause” (p. 30) – are too broad, covering both theories of interest (e.g., that the 9/11 terror attacks were orchestrated by the United States Government) and theories widely accepted as true and of no particular psychological import (that 9/11 was carried out by Al Qaeda operatives). Our interest is consequently in the former. Specifically, we define conspiracy theories as implausible explanations of significant social or political events that postulate powerful agents working together in secret to achieve a malevolent goal which often contradict the official explanations made by the relevant epistemic authority, and which are embedded into a particular tradition of explanation (Brotherton, 2013; Byford,

2011; McCauley & Jacques, 1979; Sunstein & Vermeule, 2009; Uscinski & Parent, 2014).

However, to study a construct, one first must quantify it, and existing measures have been inadequate for the task, due to problems with factor and content validity. This paper describes the development of a new scale to measure “conspiracy mentality” as an underlying trait that predicts belief in specific conspiracy theories as we have defined them, and that is distinct from “healthy scepticism,” which predicts qualitatively different kinds of conspiracies.

Conspiracy theory beliefs as a distinct construct

Conspiracy theories can be theoretically and empirically distinguished from other, superficially similar beliefs that have received attention in the literature. For example, the similarity of conspiracy beliefs and paranoid ideation led Hofstadter (1966) to identify a

“paranoid style in American politics”, and Robins and Post (1997) to discuss conspiracy theories in term of “political paranoia”. However, unlike paranoia, where the individual feels personally persecuted and interprets events as personally threatening (Feningstein & Vanable,

1992), conspiracy beliefs have an intergroup dimension (Cichocka, Marchlewska, Golec de

Zavala & Olechowski, 2015; 2016; Duran, Nicholson & Dale, 2017; Leone, Giacomantonio

& Lauriola, 2017; Mashuri & Zaduqisti, 2015; van Prooijen & van Lange, 2014). The paranoid person believes “they are out to get me”; the conspiracy theorist believes “they are out to get us.” Moreover, conspiracy theories, but not paranoia, can be considered a form of political attitude (Imhoff & Bruder, 2014). Furthermore, paranoia is related to suspicion of people in general, while conspiracy beliefs to suspicion of powerful others (Imhoff &

Lamberty, 2018). In addition, as an empirical matter, studies have found only low to moderate correlations between these two constructs (Brotherton & Eser, 2015; Darwin, Neave

& Holmes, 2011; Wilson & Rose, 2014).

Likewise, although both conspiracy theories and paranormal beliefs involve unsubstantiated and counter-normative belief systems, there are important differences between the two. For example, while conspiracy theories usually arise as a response to a conventional explanation, paranormal beliefs arise as a result of lack of conventional explanations (Wood, 2017). Furthermore, the two types of beliefs are associated with different cognitive biases; for example the conjunction fallacy predicts conspiracy beliefs, while perception of randomness predicts paranormal beliefs (Dagnall, Denovan, Drinkwater,

Parker & Clough, 2017). And while paranormal beliefs are often considered a way to resolve ambiguity, conspiracy theorists tend to endorse multiple theories about the same phenomenon, thereby increasing uncertainty about any one of them (Wood, 2017). As with paranoia, the correlation between conspiracy theories and paranormal beliefs is only moderate

(Dagnall et al., 2017; Darwin et al., 2011; Drinkwater, Dagnall & Parker, 2012; Lobato,

Mendoza, Sims & Chin, 2014; Swami et al., 2011).

Finally, and more subtly, conspiracy theories differ from real conspiracies. For example, conspiracy theories are often linked to a single, all-encompassing plot that goes far

42 beyond the immediate evil being explained and is indeed seen as a moving force in history

(Bale, 2007; Byford, 2011; Pipes, 1997). The conspirators are not merely guilty of local wrongdoing, but also of causing societal problems or committing unimaginable evil deeds

(Jane & Flemming, 2014). A conspiracy theory portrays the conspirators as monolithic, faultless and sinister agents (Bale, 2007; Pipes, 1996; Sunstein & Vermeule, 2009) whose goal is to “destroy a way of life” (Hofstadter, 1966 p.29).

In contrast, real conspiracies (e.g., bribery corruption), while often heinous, are typically limited in scope and revolve around a local issue (Bale, 2007). Furthermore, real conspiracies "involve banality and institutional disorganization" (Jane & Fleming, 2014 p.28) usually lacking in conspiracy theories. Moreover, the conspirators in real conspiracies are seen as human, with both positive and negative traits and complex, conflicting motives (Bale,

2007; Uscinski & Parent, 2014).

Previous measures of conspiracy theory beliefs

As noted, one of the defining features of individuals who believe conspiracy theories is that they tend not to believe just one. Beliefs in conspiracy theories are interrelated (e.g.

Goertzel, 1994), even if they are contradictory (Wood, Douglas & Sutton, 2012) or fictitious

(Swami et al., 2011), which suggests that a common trait, which we might call conspiracy mentality (Bruder, Haffke, Neave, Nouripanah & Imhoff, 2013), underlies beliefs in specific conspiracy theories.

The first attempts to measure this general tendency to believe in conspiracy theories consisted simply of posing a variety of conspiracy theories and asking participants to indicate their agreement or disagreement with them (e.g. Goertzel, 1994; Abalakina-Paap, Stephan,

Craig, Gregory, 1999; Swami, Chamorro-Premuzic & Furnham, 2010). There are problems with this approach, however. Although in general conspiracy beliefs are correlated, sometimes they are not (e.g. van Prooijen & Acker, 2015), and so the degree of belief

43 recorded by conspiracy theory lists depends on the particular conspiracy theories included.

Relatedly, the content specificity of these measures may limit their generalizability to certain geographical regions and/or historical periods (Brotherton, French & Pickering, 2013; Bruder et al., 2013). For example, the allegation that the Smolensk crash (Cichocka et al., 2016) was a result of a Russian conspiracy might be the issue of choice for researchers conducting their study in Poland, but not much elsewhere. A third drawback of these scales is their transparency. Participants can immediately categorize most items as conspiracy theories, which many scholars have noted have negative connotations (Bjerg & Presskorn-Thygesen,

2017; Coady, 2007a, 2007b; de Haven-Smith, 2010; de Haven- Smith & Witt, 2012) and are considered a source of stigma (Lantian et al., 2018). Thus, people will be motivated to distance themselves from conspiracy theories, resulting in socially desirable answers (but see

Wood, 2016 for findings that the labelling a statement a conspiracy theory does not affect its perceived veracity). Finally, it is unclear to what extent these scales measure the generalized tendency to believe in conspiracy theories vs. belief in specific conspiracy theories. Belief in any specific conspiracy theory is mainly driven by the underlying tendency to believe in conspiracy theories, but there are other factors that may contribute, such as political ideology

(Duran et al., 2017) or the need to feel unique (Imhoff & Lamberty, 2017). Thus, asking about a belief in specific conspiracy theories may not be a good reflection of the underlying conspiracy mentality.

Another approach to measuring conspiracy beliefs was recently suggested by Wood

(2017). Instead of measuring beliefs in specific conspiracy theories, Wood (2017) proposed measuring the underlying suspiciousness of official stories, which could be adapted to measure belief in conspiracy theories regarding almost any topic. The questionnaire consists of generic statements with blanks to be filled in with specific topics by the researcher (e.g.

The real truth about _____ is being kept from the public). For example, both someone who

44 believes that Princess Diana was murdered by the MI6 and someone who believes that she was murdered by the Royal Family are likely to agree with this statement. Thus, in this sense, Wood's scale is improvement upon previous scales as it reduces the chance that participants will be unfamiliar with a particular theory. However, the scale is nevertheless limited to measuring conspiracy thinking regarding particular events and may not be suitable for measuring conspiracy mentality per se.

In an attempt to overcome the drawbacks of the content-specific scales, researchers have begun to develop measures of generic conspiracy beliefs, without tying the beliefs to any particular content. Currently, there are two such scales available - the Generic

Conspiracist Beliefs scale (Brotherton et al., 2013) and the Conspiracy Mentality

Questionnaire (Bruder et al., 2013). The Generic Conspiracist Beliefs scale is a 15-item instrument that measures beliefs in specific conspiracy theories framed as generic statements.

For example, belief in the 9/11 conspiracy theory is assessed with the question, “The government permits or perpetrates acts of terrorism on its own soil, disguising its involvement.” The scale includes five categories of items, which Brotherton et al. (2013) named government malfeasance, extra-terrestrial cover-up, malevolent global conspiracies, personal wellbeing and control of information, with each factor represented by three items.

But despite its attempt to abstract conspiracy thinking from specific beliefs, the scale has shortcomings: its “generic” nature is belied by the existence of five content factors (whose validity has in any case been called into question; Swami et al., 2017), and by its explicit identification of the conspirators (e.g., “the government” or “scientists”). Furthermore, it is quite likely that participants see through the items’ generalized framing to map them onto salient current events.

The second measure, the Conspiracy Mentality Questionnaire, is a five-item, theoretically unidimensional measure of the general tendency to believe in conspiracy

45 theories. Unlike the Generic Conspiracist Beliefs scale, its items are truly abstract statements that do not identify a conspirator as such (e.g., “….many very important things happen in the world, which the public is never informed about”). Nevertheless, this measure, too, has shortcomings. Swami et al. (2017) have criticized some of the items for reflecting “rational beliefs about the current state of the world” (p.22) rather than conspiracy mentality. Indeed, it is very likely that the items reflect not only belief in conspiracy theories but also broader, diffuse, rational suspiciousness as well. For example, the item “Politicians usually do not tell us the true motives for their decisions”, arguably measures the awareness that a lot of political negotiations take place behind closed doors, and that the outcome of these negotiations might be “spun” when presented to the public, and in this sense it represents rational suspicion that is more or less warranted. It may be argued that the scale is unidimensional because it consistently shows high reliability (e.g. Lantian, Muller, Nurra, Douglas, 2016; Leone,

Giacomantonio, Williams & Michetti, 2018; Orosz, Krekó, Paskuj, Tóth-Király, Bőthe &

Roland-Lévy, 2016; Swami et al., 2017;). However, reliability is not an indicator of the unidimensionality of a scale (Cortina, 1993; Schmitt, 1996; Sijtsma, 2009). Indeed, using confirmatory factor analysis, Swami et al. (2017) did not find evidence that the items loaded on one factor as hypothesized. With the items of this scale supposedly measuring a one dimensional construct, but in fact reflecting more than one dimenstion, the question of construct validity is brought into question.

In sum, despite increasing interest in the nature of conspiracy theories and of the people who believe in them (e.g., Green & Douglas, 2018; Hagen, 2018; Imhoff & Lamberty,

2018; Swami, Wais, Lay, Barron & Furnham, 2016), and gains in measurement validity notwithstanding, there still exists no validated measure of the general tendency to believe in conspiracies, abstracted from both the content of specific conspiracy theories and from the

46 rational scepticism that serve democracies well. This paper presents the development of such a scale.

Study 1: Scale construction

Method

Participants. Five hundred American participants were recruited via Amazon’s

Mechanical Turk (MTurk; 187 males; 311 females; 2 “other”) and were paid 60 US cents for their participation. The mean age of the sample was M = 38.03 (range 18-75 years, SD =

12.94). The education demographics were as follows: 2.6% did not have a degree, 36.4% had high school degree, 36.6% had bachelor's degree, 14.8% had master's degree , 3.2% had doctoral degree and 6.4% had "other" educational qualifications. The number of participants was based on Field’s (2009) recommendation for at least 300 participants for a factor analysis. Thirty-five participants who failed the attention check, and two additional participants who indicated that their data were not reliable, were excluded from the analyses.

Item pool. In order to get as diverse aspects of the concept of “conspiracy theory” as possible, we asked an independent sample of 210 MTurk participants with a similar demographic profile to write their definition of conspiracy theory. Specifically, participants were told, “People have different ideas about what a conspiracy theory is. We are interested in your opinion. What is your definition of a conspiracy theory?” and were provided with text box to write their answer. There were no time or space constraints on their answers.

These 210 definitions (for full list of the definitions see Electronic Supplementary

Materials, ESM 1) were reduced to a set of 64 usable test items via the following process.

Upon careful reading and paraphrasing, 113 potential scale items were derived from the 210 definitions (e.g., “A belief that something ‘covert’ is happening behind the scenes and under

47 our noses” was converted to the item, “Something covert is happening behind the scenes and under our noses”.). Next, we screened these 113 items to identify redundancy (e.g. Major events did not happen as they are believed to vs. Important events did not happen as they are generally believed), retaining only one version of each of redundant set, leaving 64 items in the initial item pool. The items covered all three characteristics of conspiracy theories postulated by Barkun (2003): the beliefs that nothing is as it seems, that everything is connected, and that nothing happens by chance.

Procedure. The survey was presented in the online platform Qualtrics. After providing written informed consent, participants were given the following instructions:

On the next pages you will find 652 beliefs about the state of affairs in the

world. People agree or disagree to various extents with these beliefs. Indicate

your agreement or disagreement with each statement on a 1 (strongly

disagree) to 7 (strongly agree) scale. Do not overthink your answers; go with

your initial hunches.

The statements were presented in nine blocks, seven blocks consisting of seven items and two blocks consisting of eight items, including one item serving as an attention check (“Select strongly agree so that we are sure you are reading attentively”). The blocks, and the statements within blocks, were presented randomly. Finally, before being debriefed, participants were asked to assess the quality of their data with the following question “Did you honestly answered the questions in this survey? You'll be paid regardless of how you answer.”

Results

2 The 64 conspiracy theory definitions and one attention check question. 48

Factor analysis. The 64 items were subjected to Principal Component Analysis. The

Kaiser-Meyer-Olkin measure of sampling adequacy verified that the 64 items had adequate common variance for factor analysis, KMO=0.984 and all KMO values for individual items were >0.963, above the acceptable limit of 0.5 (Field, 2009). A preliminary analysis was run to explore the data with the default settings in SPSS v. 24 for extraction of factors with eigenvalues > 1. Based on this criterion five factors, explaining 66.89% of the variance, were extracted. Visual inspection of the scree plot revealed an “elbow” between the second and third factors, suggesting a two-factor solution. Thus, we conducted parallel analysis (Horn,

1965) using Stata and the package Fapara (Ender, n.d.), which confirmed that two factors should be retained (Figure 1.). Thus, we re-ran the analysis again, this time forcing a 2-factor solution and applying an oblimin rotation (=0) to improve the interpretability of the factors.

The first factor, which accounted for 1.5 times more variance than the second, was most heavily influenced by items that reflected beliefs about a malevolent agent responsible for negative events (see Table 1). Thus, we named this factor conspiracy theory ideation, as it reflects the belief that a powerful entity lies behind significant social or political events and that the conventional (official) truth is not the “real” truth. The second factor, in contrast, appeared to reflect skeptical beliefs about the limits of human knowledge, such as the fact that humans can know things only indirectly, and a more general, and arguably more rational suspicion. Thus, we named this factor skepticism.

Table 1. The Pattern Matrix of Factor Loadings for the two factors extracted with Principle Component Analysis

Component 1 2 1. A secret, covert group is responsible for tragic events. .979 -.183 2. The circumstances we are in are a result of a covert plan. .976 -.213 3. There is an unknown entity exerting secret control over events. .961 -.243 4. When one digs deep enough it is obvious that behind many events there is a conspiracy. .956 -.127 5. Many events are a result of a plot by covert groups. .949 -.114 6. The truth is known only to a secret powerful group that actively disseminates false information or misleads the .931 -.091 public. 7. There are forces at work we don’t understand that are trying to get us. .920 -.227 8. A small group of people are influencing the events taking place in the world and are trying to hide it. .845 .000 9. Many events can be explained as an “inside job” that few people know about. .842 -.004 10. The government or covert organizations are responsible for events that are unusual or unexplained. .833 -.011 11. Behind many events, there is a plan to pull something off and attempt to make it look like something else. .803 .087 12. The alternative explanations for important societal events are closer to the truth than the official story. .803 -.070 13. Many so called “coincidences” are in fact clues as to how things really happened. .797 -.013 14. The public is misled in order to hide great evil. .787 .070 15. Powerful entities are controlling matters behind the scenes. .758 .049 16. The outcomes of medical, political and social events are manipulated and controlled. .752 .105 17. Events throughout history are carefully planned and orchestrated by individuals for their own betterment. .751 .014 18. A secret of great magnitude is being kept from the public. .750 .128 19. There is secret planning and manipulation going on to make bad things happen. .739 .078 20. Powerful agents work together to suppress the truth about what really happens. .737 .170 21. Many situations or events can be explained by illegal or harmful acts by the government or other powerful .734 .076 people. 22. Some things that everyone accepts as true are in fact hoaxes created by people in power. .722 .087

23. Covert, influential individuals or organizations are responsible for many circumstances and events. .720 .179 24. We are being made to believe something other than reality. .719 .142 25. The truth regarding important events is available only to a few. .717 .120 26. Individuals or organizations are secretly operating to exert power over society on a large scale. .695 .170 27. If one digs deep enough, one will see that the commonly believed truth is wrong. .681 .178 28. Important events did not happen as they are generally believed. .679 .150 29. It is impossible that important events happen purely by chance. .659 -.198 30. Events can easily be explained by seeing how seemingly unrelated facts and coincidences are in fact linked .651 .000 together. 31. Many controversial events involve some kind of cover up. .647 .239 32. The true facts about important events are being kept secret. .643 .277 33. Oftentimes, the official explanations for events are cover-ups to hide the true cause of the event. .637 .282 34. Many illegal acts with publicly known consequences are a result of cooperation between people out of public view. .632 .167 35. Events on the news may not have actually happened. .627 .058 36. Events are portrayed to the public as one thing, when those in power know that the truth is something else entirely. .610 .280 37. Many tragic events have hidden causes behind them. .606 .236 38. Something secretive is happening right under our noses. .581 .317 39. The mainstream beliefs about history are wrong. .567 .140 40. Some events happen differently from the way scientists claim. .522 .106 41. The government is trying to control and manipulate people. .512 .357 42. People are presented with incorrect explanations for the happenings in the world. .503 .404 43. There are usually alternative explanations of why things happened. .500 .224 44. The truth is covered up by the authorities. .497 .421 45. Information is being suppressed for the benefit of a few. .489 .346 46. The real truth about many events differs from what we think is the truth. .486 .395 47. The truth differs from what the media tries to get us to believe. .486 .280 48. Important events happen as a result of more than just coincidences. .397 .370 49. All important events have some reason behind them. .360 .074

50. Many things happen without the public’s knowledge. -.072 .839 51. People try to manipulate the facts in order to suit their own agenda and motives. -.101 .783 52. There are people who don’t want the truth to come out. .035 .756 53. Information is being hidden from the public. .144 .738 54. Some things are not as they seem. .057 .701 55. Some events did not actually transpire in the way the history books tell us. .056 .676 56. People will do crazy things to cover up the truth. .089 .648 57. People are often misled into believing something that isn’t true. .258 .579 58. The government secretly conducts illegal activities. .277 .516 59. In order to find out what really happened one has to peel back the many layers of the story. .255 .506 60. Throughout history, fake public narratives have been used as explanations to hide the truth. .313 .490 61. Powerful entities like government and corporations are keeping the truth from the public for self-serving purposes. .406 .475 62. The public is being overtly lied to. .415 .474 63. Facts and events are often not as they are reported. .342 .467 64. Harmful events are being carried out by the government. .351 .429 Note: Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Items are sorted based on factor loadings. Items that loaded distinctly on their respective factors and did not load on the other factor are indicated in bold.

Figure 1. Scree plot and parallel analysis results for the original 64 items scale

Item response theory analysis. To construct a scale that would be informative across the widest possible range of  (the underlying psychological construct), we used the graded response model, suitable for polytomous (more than two response options) Likert type questions (Samejima, 1969) to examine the information function for each item, for each factor separately. Only the items that loaded distinctly on their respective factors (i.e., greater than 0.3 on conspiracy theory ideation and less than 0.1 on skepticism, and vice versa

- indicated in bold in Table 1) were analysed in IRTPRO, in order to isolate the factors as clearly as possible. Analysis of the conspiracy theory ideation subscale. None of the items exhibited local dependence (i.e., no relationship between the items existed once the underlying subconstruct was accounted for as evidenced by all LD 2 <10 (Cai, Thissen, & du Toit,

2017). The M2 goodness of fit statistic (Cai, Maydeu- Olivares, Coffman, & Thissen, 2006;

Maydeu-Olivares & Joe, 2005, 2006;) indicated some lack of fit ( M2=43.41, df=24, p=0.0091), however as RMSEA indicated good fit (RMSEA=0.04), the M2 value may be due to a limited amount of model error inherent in any model (Cai et al., 2017).

However, as evidenced by the S-2 item-fit statistic (Orlando, & Thissen, 2000,

2003) one item demonstrated misfit at p<0.05 after controlling for type 1 error with the

Benjamini-Hochberg adjustment (Benjamini & Hochberg, 1995). After dropping this item, all items fitted the model, no item exhibited local dependence (all LD 2 <10) and the model had good fit (M2=26.97, df=15, p=0.02, RMSEA=0.04).

Next, we examined the item information function values to select the items that would comprise the final subscale. Because this full subscale provided the least information at the low and high end of the underlying construct (see Table 2), in the selection process we gave priority to items that provided high information at the ends: at least 0.3 at both 2.8SD of .

Based on these criteria items #10, 12, 13, 17, 21, 22, 30 and 35 were selected. Item 30

54 however, provided low information overall and its content was substantially overlapping with the content of item 13. Thus, the final subscale consisted of seven items (see Table 1 in ESM

3).

Analysis of the skepticism subscale. No pair of items exhibited local dependence as evidenced by all the LD 2 <10 . In addition, the model was good fit for the data

(RMSEA=0.04). After the Benjamini-Hochberg adjustment was applied item 55 exhibited model misfit as evidenced by the S-2 . Thus, the procedure was run again. The revised model exhibited good fit (M2=363.212, df=212, p=0.0001, RMSEA=0.04), and as the four items exhibited good fit it was decided to retain them all. Although the range of discrimination for this subscale is far from perfect (see Table 3), it should be noted that the nature of this dimension may be such that it may not be possible to make fine distinction at the high level of the dimension.

Table 2. Item Information Function Values at 15 Values of θ from -2.8 to 2.8 for conspiracy theory ideation

θ Item # -2.8 -2.4 -2.0 -1.6 -1.2 -0.8 -0.4 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 Item 6 0.03 0.12 0.51 1.76 3.53 3.88 4.16 4.23 3.92 3.90 3.89 3.34 1.49 0.42 0.10 Item 8 0.11 0.42 1.34 2.84 3.33 3.52 3.64 3.58 3.29 3.35 3.23 2.78 1.29 0.39 0.11 Item 9 0.23 0.71 1.75 2.64 2.70 2.95 3.05 3.00 2.84 2.82 2.68 2.56 1.54 0.59 0.19 Item 10 0.36 1.00 2.06 2.52 2.56 2.76 2.81 2.79 2.64 2.63 2.60 2.56 1.96 0.92 0.32 Item 11 0.20 0.77 2.24 3.43 3.34 3.77 3.90 3.89 3.66 3.70 3.59 2.76 1.09 0.30 0.07 Item 12 0.52 0.94 1.33 1.48 1.53 1.57 1.56 1.54 1.53 1.52 1.51 1.47 1.39 1.08 0.63 Item 13 0.45 0.93 1.46 1.67 1.73 1.83 1.87 1.84 1.76 1.74 1.70 1.67 1.51 1.00 0.50 Item 14 0.22 0.62 1.43 2.19 2.33 2.46 2.55 2.59 2.50 2.48 2.44 2.20 1.37 0.58 0.20 Item 15 0.46 1.06 1.79 2.10 2.18 2.25 2.26 2.12 2.09 2.13 2.07 1.64 0.89 0.37 0.13 Item 17 0.46 0.92 1.40 1.58 1.62 1.71 1.76 1.76 1.69 1.66 1.68 1.64 1.43 0.94 0.47 Item 19 0.21 0.54 1.17 1.81 2.00 2.10 2.20 2.20 2.16 2.12 2.07 1.97 1.48 0.77 0.31 Item 21 0.56 1.23 1.90 2.04 2.14 2.27 2.26 2.22 2.15 2.16 2.09 2.04 1.58 0.82 0.33 Item 22 0.30 0.69 1.28 1.73 1.88 1.94 1.96 1.92 1.87 1.84 1.77 1.72 1.36 0.77 0.34 Item 30 0.72 0.84 0.89 0.91 0.93 0.95 0.95 0.95 0.94 0.94 0.93 0.91 0.86 0.74 0.53 Item 35 0.33 0.54 0.76 0.91 0.96 0.99 1.00 1.01 1.01 1.00 1.00 0.96 0.84 0.63 0.40 Item 49 0.24 0.24 0.24 0.24 0.24 0.24 0.23 0.23 0.23 0.23 0.23 0.22 0.22 0.21 0.19

Test Information: 6.38 12.57 22.55 30.87 34.01 36.16 37.15 36.89 35.29 35.21 34.48 31.42 21.29 11.51 5.84 Expected s.e.: 0.40 0.28 0.21 0.18 0.17 0.17 0.16 0.16 0.17 0.17 0.17 0.18 0.22 0.29 0.41

Marginal Reliability for Response Pattern Scores: 0.98

Table 3. Item Information Function Values f at 15 Values of θ from -2.8 to 2.8 for skepticism

θ Item # -2.8 -2.4 -2.0 -1.6 -1.2 -0.8 -0.4 -0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 Item 50 1.91 2.03 2.08 2.08 1.96 1.89 1.85 1.72 1.78 1.53 0.89 0.39 0.15 0.06 0.02 Item 52 1.68 2.06 2.15 2.12 1.95 1.80 1.90 1.90 1.87 1.56 0.89 0.39 0.15 0.05 0.02 Item 54 1.68 2.11 2.19 2.13 1.88 1.77 1.88 1.76 1.77 1.75 1.20 0.57 0.23 0.08 0.03 Item 56 1.53 1.70 1.73 1.71 1.62 1.52 1.55 1.52 1.50 1.41 1.00 0.54 0.24 0.10 0.04

Test Information 7.78 8.89 9.14 9.05 8.42 7.98 8.18 7.90 7.92 7.25 4.99 2.90 1.77 1.30 1.11 Expected SE 0.36 0.34 0.33 0.33 0.34 0.35 0.35 0.36 0.36 0.37 0.45 0.59 0.75 0.88 0.95 Marginal Reliability for Response Pattern Scores: 0.84

Discussion

We set out to develop a scale for measuring conspiracy mentality, a general approach to the world in which significant social or political events are explained as the workings of a powerful agents with malevolent goal, and the official truth is regarded as an illusion. By factor analyzing a large set of spontaneously generated descriptions of conspiracy theories, we abstracted one primary factor that appears to capture this construct. A second factor also emerged, which appeared to tap into a more rational or healthy skepticism about the world.

Studies 2 and 3 aim to distinguish the two dimensions empirically and to provide data on their validity. In particular, if conspiracy theory ideation is distinct from skepticism, the two dimensions should predict beliefs in qualitatively different kinds of claims (once shared variance is accounted for), with the former associated with traditional conspiracy theories, and the latter with plausible allegations (e.g., of bribery or corruption).

Study 2. Predictive validity and confirmatory factor analysis

Method

Participants. Two hundred and two American MTurk participants filled an online survey in exchange for 50 US cents. Here and throughout the thesis none of the participants have taken part in previous studies reported in the thesis, unless explicitly stated. The mean age of the sample was 34.75, (SD=11.92; range 19 – 70 years). The other demographics were as follows: 108 said they were female, 93 male, 1 other; 2% did not have a degree, 31.2% had high school degree, 45% had bachelor's degree, 10.4% had master's degree, 3.5% had doctoral degree and 7.9% reported "other" as their highest degree. To detect a significant slope of B=0.2 with 80% power at alpha 0.05 we needed 191 participants. We slightly over recruited to allow for attention check failures.

Materials.

Stimulus development. The conspiracy theories used in this study were developed in two phases. In the first phase 210 MTurk participants were asked to write five conspiracy theories they have read or heard about, from which we extracted 112 unique conspiracy theories. In the second phase these 112 conspiracy theories were given to a new sample of

MTurk participants (N=233) who were asked to rate each theory in terms of "how good an example" it was of a conspiracy theory, using a 9 point scale. The 15 items with the highest prototypicality ratings were planned to be included as stimuli, however due to an administrative error when creating the survey the 15th conspiracy theory was accidentally replaced by another (ranked 36th). We included this conspiracy theory in all further analysis.

The mean prototypicality rating for these fifteen statements was 6.36.

The corruption and bribery scenarios were taken and adapted from the United States

Internal Revenue Service’s Examples of Public Corruption Investigations for the fiscal year

2016 and 2015 (Internal Revenue Service, 2017a, Internal Revenue Service, 2017b). By choosing scenarios of bribery and corruption we ensured that the statements in question would involve conspiracy by definition. The complete list of statements used is given in

Electronic Supplementary Materials 2.

For the conspiracy theories and corruption and bribery scenarios the participants needed to indicate how likely each is on a 7 – point scale (1=Very unlikely, 7= Very likely).

Procedure. Participants completed an online questionnaire in Qualtrics in return for a payment of US$0.50. The conspiracy theory and corruptions statements were presented in one block in randomized order. The conspiracy mentality scale developed in Study 1 was presented in a separate block, with the order of presentation of the two blocks counterbalanced. Within each block there was one attention check question.

Results and Discussion

Seventeen participants failed one or both of the attention check questions, leaving 185 participants for analysis.

Confirmatory factor analysis. Using AMOS v. 23 and the maximum likelihood method we run confirmatory factor analysis for the scale, allowing the two factors to correlate. All items loaded on the respective factors as hypothesized (see Table 1 in ESM 3).

In addition, the two-dimensional model was good fit for the data (χ2/df =1.87, CFI=0.972,

TLI= 0.964, RMSEA=0.069, PCLOSE=0.092).

Predictive validity. An exploratory factor analysis of the conspiracy theories and the corruption cases initially indicated the presence of four factors, but parallel analysis suggested only two. Furthermore, the two factor solution showed distinct loadings for the conspiracy theory statements on one factor and corruption cases on the other (reliability coefficients = .93 and .94 respectively). Thus, we decided to analyze a composite (average) score for both types of statements. In addition, we ran separate regressions to test whether the pattern of results we hypothesized would hold even for individual items.

Results for the composite score as dependent variable. When conspiracy theory ideation and skepticism were simultaneously entered as predictors and the composite score for the conspiracy theories as dependent variable, the model explained 67% of the variance.

However only conspiracy theory ideation was significant predictor (B=0.832, SE=0.055, p

<0.0001); skepticism was not (B=0.051, SE=0.076, p = 0.50).

When the analysis was repeated with the composite score for the corruption cases instead of the conspiracy theories as a dependent variable, the opposite pattern emerged. The model could explain 15% of the variance. Conspiracy theory ideation was not significant predictor (B=0.070, SE=0.062, p=0.26), but skepticism was (B=0.331, SE=0.086, p<0.001).

Results for the individual items as dependent variable. Thirty one multiple regressions were run. In each conspiracy theory ideation and skepticism were entered as a predictor variable and one of the conspiracy theories or corruption cases as outcome variable.

The results for the conspiracy theories are presented in Table 4. and for the corruption allegations in Table 5.

Table 4. Unstandardized coefficients, standard errors, standardized coefficients and p values for the conspiracy theories as outcome variables.

Conspiracy theory ideation Skepticism 2 BCM SE βCM p (corrected p) BSK SE βSK p (corrected p) R 1. Princess Diana's death 0.870 0.109 0.583 0.000 (0.000) 0.059 0.151 0.028 0.698 (0.73) 0.360 2. Murder of JFK 0.656 0.107 0.450 0.000 (.000) 0.416 0.148 0.207 0.005 (0.012) 0.356 3. Aliens in Roswell 0.802 0.122 0.506 0.000 (0.000) 0.128 0.169 0.058 0.450 (0.55) 0.294 4. Suppressed cures 0.927 0.118 0.573 0.000 (0.000) 0.121 0.163 0.054 0.460 (0.55) 0.368 5. Marilyn Monroe's death 0.798 0.108 0.544 0.000 (0.000) 0.169 0.149 0.083 0.259 (0.363) 0.357 6. The Clintons killed people 0.933 0.118 0.576 0.000 (0.000) 0.118 0.163 0.053 0.470 (0.56) 0.370 7. 2017 elections rigged 0.430 0.132 0.275 0.001 (0.003) 0.345 0.182 0.160 0.059 (0.11) 0.153 8. TWA Flight 800 shot down 0.926 0.088 0.721 0.000 (0.000) -0.199 0.122 -0.112 0.105 (0.179) 0.438 9. Cameras on phones spy citizens 0.799 0.107 0.557 0.000 (0.000) 0.067 0.148 0.034 0.650 (0.707) 0.334 10. Powerful, secret elite rules 1.089 0.094 0.740 0.000 (0.000) -0.085 0.130 -0.042 0.513 (0.589) 0.513 11. Government kills people 0.890 0.091 0.645 0.000 (0.000) 0.155 0.125 0.081 0.219 (0.323) 0.484 12. Government involved in 9/11 0.918 0.103 0.638 0.000 (0.000) -0.042 0.142 -0.021 0.766 (0.77) 0.391 13. Media used to control minds 0.916 0.110 0.593 0.000 (0.000) 0.138 0.151 0.065 0.362 (0.464) 0.401 14. Faked moon landing 0.710 0.099 0.574 0.000 (0.000) -0.288 0.136 -0.169 0.036(0.074) 0.243 15. Sandy Hook shootings faked 0.820 0.098 0.642 0.000 (0.000) -0.335 0.135 -0.190 0.014(0.032) 0.304 Note: B= unstandardized coefficients, β= standardized coefficients.

Table 5. Unstandardized coefficients, standard errors, standardized coefficients and p values for the corruption cases as outcome variables.

Conspiracy mentality Skepticism 2 Statement BCM SE βCM p (corrected p) BSK SE βSK p (corrected p) R 1. City council official abuses power 0.088 0.086 0.090 0.309 (0.407) 0.296 0.119 0.218 0.014 (0.032) 0.079 2. CEO bribes UN official 0.127 0.090 0.121 0.159 (0.243) 0.400 0.124 0.276 0.001 (0.003) 0.130 3. Executive director abuses power 0.045 0.088 0.046 0.606 (0.670) 0.323 0.121 0.236 0.008 (0.01) 0.071 4. School principal receives kickbacks 0.169 0.106 0.143 0.111 (0.181) 0.157 0.146 0.097 0.282 (0.38) 0.046 5. Employee solicits money 0.032 0.083 0.035 0.697 (0.733) 0.254 0.114 0.199 0.027 (0.059) 0.049 6. A man pays to conceal misconduct -0.049 0.081 -0.053 0.547 (0.616) 0.419 0.112 0.328 0.0002 (0.0006) 0.090 7. President of Airport Authority abuses power 0.112 0.100 0.100 0.264 (0.363) 0.270 0.138 0.173 0.053 (0.106) 0.060 8. Senator abuses power -0.081 0.089 -0.079 0.367 (0.464) 0.478 0.123 0.339 0.0001 (0.0003) 0.090 9. A man solicits kickbacks 0.028 0.087 0.028 0.747 (0.77) 0.370 0.120 0.273 0.002 (0.005) 0.084 10. Building Manager abuses power 0.024 0.081 0.026 0.770 (0.77) 0.204 0.111 0.165 0.068 (0.124) 0.033 11. A director bribes officials 0.147 0.083 0.150 0.080 (0.14) 0.352 0.115 0.261 0.003 (0.008) 0.137 12. An employee offers favor for cash 0.189 0.103 0.162 0.068 (0.12) 0.230 0.142 0.143 0.107 (0.179) 0.074 13. Utility authority boards accepts bribes -0.061 0.093 -0.057 0.511 (0.589) 0.510 0.129 0.345 0.0001 (0.0003) 0.099 14. A chief of staff abused his position 0.147 0.101 0.128 0.147 (0.233) 0.303 0.139 0.191 0.031 (0.066) 0.082 15. A former official abused his position 0.086 0.074 0.098 0.248 (0.357) 0.377 0.102 0.313 0.0002 (0.0006) 0.144 16. A director accepts money 0.114 0.081 0.121 0.161 (0.243) 0.344 0.112 0.265 0.002 (0.005) 0.122 Note: B= unstandardized coefficients, β= standardized coefficients.

As can be seen from Table 3. and Table 5., after applying the Benjamini-Hochberg adjustment, the conspiracy theory ideation predicted beliefs in conspiracy theories, but not belief in the corruption cases, while the skepticism predicted belief in most of the corruption cases, but positively predicted only one conspiracy theory.

In sum, Study 2 confirmed the factor structure uncovered by our initial analysis of the scale items and, equally important, illustrated the predictive and discriminant validity of the two factors. A challenge for social scientists has been identifying an underlying construct that predicts the particular category of beliefs we know as conspiracy theories, as opposed to beliefs that also qualify as conspiratorial claims but would be considered rational or plausible. The difference, in theory, is not merely in what is socially acceptable; conspiracy theories have distinct assumptions implicit in them, such as a Manichean worldview of struggle between good (the people) and evil (the conspirators), (Bale, 2007; Barkun, 2003;

Oliver & Wood, 2014; Cubitt, 1989; 1993; Brotherton, 2015), conceiving the conspirators as omnipotent, omnipresent and monolithic (Bale, 2007) or seeing the conspiratorial plot as all- encompassing and the moving force in history (Bale, 2007; Byford, 2011; Pipes,1997). Our conspiracy theory ideation factor appears to tap into those assumptions, and not into the rational skepticism that underlies beliefs or suspicions about actual conspiracies or conspiracies more generally, while our skepticism factor does the opposite. At the same time, it does not appear that the skepticism factor adds predictive validity, so while it may well be useful for assessing doubt and suspicion regarding mundane abuses of power, its usefulness for predicting attitudes and behavior toward specific conspiracy theories may be limited.

Study 3a. Convergent and divergent validity

Overview and rationale

If our two factors do indeed measure and distinguish conspiracy theory ideation and skepticism, they should be associated with other established measures that relate to these constructs, and also unassociated with others that are theoretically independent. In particular, as argued in the Introduction, we expected conspiracy theory ideation to be positively related to paranormal beliefs and paranoid ideation and, based on previous work (Goertzel, 1994,

Abalakina – Paap et al., 1999, Mashuri & Zaduqisti, 2014) to be negatively related to trust, and independent of death anxiety (Bruder et al., 2013) and emotional intelligence (Brotherton et al., 2013). Research on personality and conspiracy theory beliefs (Swami & Furnham,

2012; Swami et al., 2010, 2011, 2013) also suggests a positive relation with openness to experience, a negative relation with agreeableness, and no relation with conscientiousness, neuroticism and extraversion. Skepticism, on the other hand, should be correlated with rational thinking style (as opposed to conspiracy theory ideation which previous studies have found to be positively correlated with intuitive thinking style and negatively with analytic thinking style (e.g. Swami, Voracek, Stieger, Tran & Furnham, 2014) and with other measures of skepticism such as skepticism towards advertisement and electronic word of mouth skepticism.

We also included a new set of specific conspiracy theories to replicate the predictive validity findings from study 2 in relation to specific conspiracy theories.

Method

Participants. Three hundred eight American MTurk participants (189 female, 118 male, 1 “other”) filled in an online survey in Qualtrics in exchange for US$ 1.00. The mean age of the sample was M=36.30 years ( SD=11.52, age range 18 to 77 years). The education

65 demographics were as follows: 0.6% did not have a degree, 27.6% had high school degree,

46.8% had bachelor's degree, 18.2% had master's degree, 3.6% had doctoral degree and 3.2% selected "other" as their highest degree. To detect a significant slope of B=0.2 with 80% power at alpha 0.05 we needed 191 participants. We significantly overrecruited, anticipating a number of attention check failures due to the length of the questionnaire.

Measures and procedure. After providing informed consent, participants were presented with the instruments in the following order.

Personality was assessed on the ten item personality inventory (TIPI), used and validated in many previous studies (e.g. Gosling, Rentfrow & Swann, 2003). Participants rate the degree to which each of five pairs of traits (two for each of the Big Five dimensions, with one of each pair reverse scored) applies to them, using a 7-point scale anchored at “disagree strongly” and “agree strongly.” The scores from each pair of items are averaged to arrive at a single score for each dimension. Higher scores indicate higher presence of the trait.

Trust was assessed using three items previously used by Goertzel (1994) in research on beliefs in conspiracy theories. Participant rated their agreement or disagreement on a five - point scale (1=strongly disagree, 5=strongly agree) with statements about whether they can trust the police, their neighbors or their relatives. Answers across the three items were averaged to arrive at a single score.

Paranormal beliefs were measured on the Revised Paranormal Beliefs Scale (R-PBS;

Tobacyk, 2004), a 26-item scale that measures seven dimensions of paranormal beliefs: traditional religious beliefs, psi, witchcraft, superstition, spiritualism, extraordinary life forms and precognition. Participant rate their agreement with statements relating to each dimension, using a 7 point scale anchored at 1 (=strongly disagree) to 7 (= strongly agree). Scores for

66 each subscale are averaged (one item from the psi subscale is reverse coded) with higher scores indicating stronger beliefs on a dimension.

Belief in specific conspiracy theories was measured on the 15-item Conspiracy

Theories Belief Inventory (CTBI; Swami, et al., 2010; Swami et al., 2011). Participants rate their agreement with specific conspiracy theories [e.g., "The Coca Cola company intentionally changed to an inferior formula with the intent of driving up demand for their classic product, later reintroducing it for their financial gain", 1 (= completely false), 9 (= completely true)], and a total score is calculated by averaging ratings across all items, with higher scores indicating stronger overall belief in the theories.

Death anxiety was measured with the Death Anxiety Questionnaire (DAQ; Conte,

Weiner & Plutchik, 1982), a 15 item scale that measures attitudes towards death and dying, including subfactors relating to fear of the unknown, fear of suffering, fear of loneliness and fear of extinction. Participants rate their agreement with statements relating to each dimension using a three point scale anchored at 0 = not at all and 2 = very much. Scores are averaged across items to arrive at a total score, with higher scores indicating higher death anxiety.

Paranoid ideation was measured with the Paranoia scale (Feningstein & Vanable,

1992), a 20-item measure that was designed to assess paranoid thinking in non-clinical population (e.g., “Some people have tried to steal my ideas and take credit for them”).

Participants rate the degree to which each statement applies to them on a 5-point scale anchored from 1 (= not at all applicable to me) to 5 (= extremely applicable to me). The answers are averaged across the items to obtain a total score with higher scores indicating higher paranoid ideation.

Analytic thinking style was measured with the analytic subscale of the Rational-

Experiential Inventory (REI-40) developed by Pacini and Epstein (1999). Its questions tap into rational ability (e.g. “I am much better at figuring things out logically than most people”) and rational engagement (e.g. “I enjoy thinking in abstract terms”). Participants rate the degree to which the items are true of themselves on a 5 point scale anchored from 1 (= definitely not true of myself) to 5 (= definitely true of myself). After reverse scoring several items, scores are averaged across all items with higher scores indicated greater rational ability and engagement.

Skepticism was assessed with two instruments. First, the 9-item Electronic Word of

Mouth (eWOM) (Zhang, Ko & Carpenter, 2016) was used to measure three dimensions of skepticism about online claims: truthfulness (e.g. “We can hardly depend on getting the truth from most online reviews”), motivation (e.g. “Online reviewers care more about getting you to buy things”), and identity (e.g. “People writing online product reviews are not necessarily the real customers”). Participant rate their agreement with statements relating to each dimension on a 7 point scale anchored from 1 (= strongly disagree) to 7 (= strongly agree).

The answers are averaged across the items to obtain total score with higher scores indicate greater levels of skepticism.

Second, the nine item SKEP scale (Obermiller & Spangenberg, 1998) was used to measure skepticism toward advertising claims (e.g., "We can depend on getting the truth in most advertising”). Participants rate their agreement with each statement on a five point scale, anchored from 1 (= strongly agree) to 5 (= strongly disagree). Answers are averaged across all items with higher scores indicating greater skepticism.

Finally, emotional intelligence was measured with the 16 item Wong-Law Emotional

Intelligence Scale (Wong & Law, 2002) developed to measure four facets of emotional intelligence: self emotional appraisal (e.g. "I always know whether or not I am happy"),

68 other’s emotional appraisal (e.g. "I always know my friends’ emotions from their behavior"), regulation of emotion (e.g. "I can always calm down quickly when I am very angry") and use of emotion (e.g. "I always set goals for myself and then try my best to achieve them").

Participants rate their agreement on a 7- point scale anchored from 1 (= strongly disagree), to

7 (= strongly agree). Scores are averaged across items, with higher scores indicating greater emotional intelligence.

One attention check item was included within each of the following scales: skepticism subscale of the Conspiracy Mentality scale ("To make sure you read attentively please select strongly agree"), Revised Paranormal Beliefs Scale ("If you read attentively select strongly agree"), Paranoid ideation scale ("I am reading attentively - select extremely applicable to me if you read this") and SKEP ("To make sure you read attentively select strongly agree").

Failure to select the requested response on any of these was grounds for exclusion from the analyses.

Results and Discussion

Seventy-nine participants failed one or more of the attention checks, leaving 229 participants for analysis.

A series of multiple regressions, using the conspiracy theory ideation and skepticism dimensions simultaneously to predict each dependent variable in turn, were used to assess convergent and divergent validity. The Benjamin-Hochberg correction was used to keep the overall type 1 error rate to .05. The results are shown in Table 6.

Table 6. Unstandardized coefficients, standard errors, standardized coefficients and p values for the examined variables.

Cronb Conspiracy theory ideation Skepticism ach α Dependent variable B SE ß p (corrected p) Expected B SE ß p (corrected p) Expected relationship relationship Paranormal beliefs 0.95 0.473 0.067 0.483 0.000 **(0.000**) positive 0.032 0.074 0.029 0.671 (0.782) N/A Paranoid ideation 0.92 0.163 0.043 0.284 0.000** (0.000**) positive 0.053 0.047 0.083 0.264 (0.462) N/A Trust 0.62 -0.097 0.043 -0.173 0.026* (0.091+) negative -0.024 0.048 -0.039 0.613 (0.782) N/A Death anxiety 0.86 -0.004 0.025 -0.014 0.863 (0.87) none 0.012 0.028 0.035 0.657 (0.782) N/A Rational thinking style 0.93 -0.111 0.043 -0.198 0.011* (0.0513+) negative 0.099 0.048 0.159 0.040* (0.101) positive Ad skepticism 0.94 -0.126 0.053 -0.183 0.018* (0.072+) N/A 0.158 0.059 0.207 0.008** (0.044*) positive eWOM skepticism 0.93 0.212 0.069 0.230 0.002** (0.014*) N/A 0.094 0.077 0.092 0.222 (0.414) positive Emotional intelligence 0.92 -0.024 0.052 -0.036 0.641 (0.782) none 0.125 0.057 0.169 0.030* (0.093+) N/A Extraversion 0.76 0.059 0.101 0.046 0.561 (0.782) none -0.067 0.112 -0.047 0.549 (0.782) N/A Agreeableness 0.60 -0.024 0.078 -0.024 0.762 (0.827) negative 0.131 0.087 0.118 0.132 (0.308) N/A Conscientiousness 0.62 0.013 0.078 0.013 0.870 (0.870) none 0.069 0.087 0.062 0.428(0.704) N/A Emotional stability 0.76 0.026 0.089 0.023 0.768 (0.827) none -0.065 0.099 -0.052 0.511(0.782) N/A Openness to experience 0.57 0.105 0.078 0.103 0.180 (0.387) positive 0.185 0.087 0.162 0.034* (0.095+) N/A Belief in specific CT 0.93 0.884 0.074 0.666 0.000** (0.000**) positive 0.102 0.082 0.069 0.216 (0.414) N/A Note: ** p<0.01, *p<0.05, + p<0.1 ., eWOM= electronic word of mouth, B= unstandardized coefficient, β= standardized coefficient.

The results are summarized in Table 6. As predicted participants scoring higher on the conspiracy theory ideation scale also tended to hold paranormal and paranoid belief and to endorse specific conspiracy theories. Further, as expected, there was no relationship between our measure of conspiracy theory ideation and death anxiety, emotional intelligence, conscientiousness, neuroticism and extraversion. Also, the scale’s relationship with trust and rational thinking style were marginally significant (p<0.1) and in the expected (negative) direction.

We did not find any relationship between conspiracy theory ideation and openness to experience or agreeableness, perhaps not surprisingly given that the correlation in previous studies has been rather low (e.g. Swami et al, 2011), inconsistent (e.g., Galliford & Furnham,

2017; Imhoff & Bruder, 2014), and may stem due to shared variance with skepticism and not conspiracy theory ideation per se.

Moving on to skepticism, as expected, there was a positive relationship between our measure of skepticism and ad skepticism and a marginal relationship with rational thinking style. However, while we expected positive relationship with electronic word of mouth skepticism, we did not find support for this hypothesis. There was also a marginally positive relationship with emotional intelligence and openness to experience. Furthermore, there was no association with paranormal beliefs, paranoid ideation or trust.

Unexpectedly, some measure predicted to correlate with the skepticism dimension were also associated, inconsistently, with conspiracy theory ideation (word of mouth positively, and advertisement skepticism negatively). While the negative correlation with advertisement skepticism is less surprising, the correlation with electronic word of mouth skepticism is more puzzling. One speculative account is that the focus and the prescribed agency in the items of the eWOM scale might have presented the claims in a way that easily

71 tap into conspiracy theory ideation. For example "people writing online reviews are not necessarily the real customers" perhaps implies a larger ongoing plot.

Study 3b. Test – Retest reliability

All participants in Study 3a were invited to complete the scale again (along with a second, unrelated measure that appeared after it) 11 weeks after their initial participation. The survey was open for two weeks, by which point 109 participants (71 females, 38 males, mean age at T2= 41.11, range 20 to 74) had taken part. After the participants that failed the attention check at either time 1 or time 2 were removed, the final sample consisted of 86 participants (56 females, 30 males, Mage 39.97, range 18-74). The correlation between the conspiracy theory ideation factor at time 1 and time 2 was r=0.714, p<0.001 (one tailed), and between the skepticism factor r=0.710, p <0.001 (one tailed). Paired samples t-tests showed no significant difference between time 1 and 2 for the conspiracy theory ideation factor t(84)=1.204, p=0.232 (Mtime1= 3.96, SDtime1=1.32; Mtime2=4.08, SDtime2=1.17 or for the skepticism factor t(84)=-0.062, p=0.951 (Mtime1=5.70, SDtime1=1.23 Mtime2=5.70, SDtime2=1.03.

Thus, the results indicate that the scale can be used as a reliable measure of conspiracy theory ideation and skepticism.

Study 4. Construct validity across three nations

In order to demonstrate construct validity beyond MTurk samples, in Study 4 we collected data from two additional countries, New Zealand and Macedonia, focussing specifically on the invariance of the factor loadings and the underlying latent structure.

Method

Participants and procedure. The New Zealand sample consisted of 132 students (27 males, 102 females, 1 other; 2 did not disclose their gender) who participated in exchange for extra course credit. Participants filled in the questionnaire on a computer running PsychoPy

72 software (Peirce, 2007). Up to four participants were tested in a single session. Participants were seated in the same laboratory room but separated by dividers.

The Macedonian participants comprised a representative sample (N=311; 145 males,

166 females). Data were collected by a local market research company, along with other measures for an unrelated correlational study. The questionnaire was in Macedonian and the scale was translated into Macedonian by two individuals fluent in both languages.

Disagreements were settled by discussion about the most appropriate wording. The conspiracy mentality scale was administered last.

A new sample of MTurk participants (N= 357; 168 males, 188 females, 1 other) was pooled from three different experimental studies. Only the data from the participants that were allocated to the control group in these studies was used.

None of the participants had been subject to any effective experimental manipulation.

Results and Discussion

To test for the factorial equivalence of scores from the Conspiracy Mentality Scale we used AMOS v. 23 and followed the procedure outlined by Byrne (2009). We first established configural invariance by demonstrating that the items loaded on the respective factors in the three groups as expected (see Table 2, in Electronic Supplementary Materials 3). The results indicated that the two-factor structure is excellent fit for the data across the three groups (see

Table 7.).

Next, we tested for metric invariance by constraining the regression weights to be equal across the groups, and compared the constrained model fit indices with the ones from the configural model. A significant Δχ2 indicates nonequivalence. Given that, Δχ2 is dependent on sample size and in large samples is almost always significant, changes in other fit indexes were also considered: ΔCFI and ΔTLI greater than 0.01 (Cheung & Rensvold,

2002), ΔRMSEA values greater than 0.015 (Chen, 2007) indicate measurement

73 nonequivalence. As can be seen from Table 7. all indices apart from the Δχ2 indicated metric equivalence.

Finally, we tested for structural invariance by constraining the factor variance and covariances to be equal and comparing this constrained model fit to the configural model.

Again, all indices except Δχ2 argued for invariance.

Thus, we concluded that across the three cultural contexts examined, there is support for cross-cultural equivalence of the relations among the factors of the Conspiracy Mentality

Scale, as well as between each factor and the respective indicators.

Table 7. Results from test of the factorial equivalence of scores from the Conspiracy Mentality Scale.

Model χ2 RMSEA (90% Δχ2 ΔRMS Model CFI TLI comparis ΔCFI ΔTLI Decision (df) CI) (Δdf) EA on M1: 309.400 0.964 0.042 0.954 - - - - -

Configural (129) (0.036 - 0.048) invariance

M2: Metric 350.632 0.960 0.042 0.955 M1 41.232(18) 0.004 0 0.001 Equivalence invariance (147) (0.036-0.047)

M3: 380.905 0.955 0.043 (0.038- 0.951 M1 74.251 (4) 0.009 0.003 0.002 Equivalence

Structural (153) 0.049) equivalence

Note: CFI= Comparative Fit Index, RMSEA= Root mean square error of approximation, TLI= Tucker Lewis Index.

General discussion

Studying the psychological underpinnings of belief in conspiracy theories is important because they are associated with negative social, health and political outcomes. Previous measures of beliefs in conspiracy theories have either focused on belief in specific conspiracy theories or have measured the general tendency to believe in conspiracy theories. However, the first approach has limited use and may not adequately capture the underlying construct, while existing scales taking the second approach have either not been sufficiently generic, or have confounded the tendency to believe in conspiracy theories with other variables such as knowledge about the current state of the world. Our goal in this paper was to develop and validate a scale that measures the general tendency to believe in conspiracy theories and to improve on previous efforts.

In Study 1 we explored the factor structure of the initial item pool, generated by 210

MTurk participants who gave their definition of a conspiracy theory, and discovered two factors, which we termed conspiracy theory ideation and skepticism. After maximizing the information value of each dimension, we settled on an 11-item scale, which subsequently showed good test- retest reliability, as well as convergent, divergent, and predictive validity. In particular, conspiracy theory ideation correlated positively with paranoia, paranormal beliefs, and marginally negatively with trust, and, equally important, did not correlate with emotional intelligence, death anxiety, extraversion, consciousness, or emotional stability. Skepticism, meanwhile, correlated with advertisement skepticism. Further, once the shared variance was accounted for, both dimensions predicted different types of conspiracies. Finally, in study 4 we demonstrated construct validity in two other national contexts.

We argue that these two dimensions represent two sides of the conspiracy mentality construct (the general tendency to believe in any conspiracy)– the “rational” and the “irrational”, which become especially evident once their shared variance is controlled. We have evolved to be suspicious because it is adaptive – higher sensitivity to deception means that we are less likely to miss potential threats (Kreko, 2015) and this suspiciousness is captured by the “rational side” of conspiracy mentality – our skepticism dimension. Meanwhile, the “irrational side” reflects the point at which rational suspicion morphs into nihilism and becomes maladaptive, for example by dedicating one's life to "uncovering" the conspiracy or enlightening the "sheeple".

We acknowledge that our treatment of conspiracy theories, and particularly our distinction between plausible and implausible theories, involves some subjectivity. Who are psychologists to judge what is plausible or rational to believe? And haven’t some “implausible” theories, such as allegations of widespread government surveillance, turned out to be true?

Fortunately, the plausibility of a particular claim is also an empirical matter, and can be quantified as a claim’s acceptance in the population. That acceptance may well change over time as evidence accumulates or social attitudes shift, and if they do, some conspiracy theories will no longer qualify as such, and new, perhaps contradictory conspiracy theories will emerge. (Today, someone who denies government surveillance has a better claim to the title of “conspiracy theorist.”) However, the fact that norms change does not prevent a meaningful distinction between beliefs that contradict those norms, and those that are consistent with them.

We recommend the use of the skepticism dimension as a control variable in situations where researchers wish to establish the unique relationship between conspiracy theory ideation and a construct of interest, in order to distinguish such ideation from general skepticism that is not specific to conspiracy theory beliefs. For example, in Study 3a the relationships between

conspiracy theory ideation and some variables (openness to experience, trust) was accounted for by shared variance with skepticism; failure to control for skepticism in these cases would lead to spurious conclusions about conspiracy theorists’ beliefs. In addition, our skepticism dimension may prove useful in its own right. Current literature on skepticism as a psychological construct is fractionized in specific areas. For example, academics are interested in auditors’ professional skepticism (Hurtt, 2010), consumer skepticism (Ford, Smith & Swasy, 1990; Forehand & Grier,

2003), investor skepticism (Demerens, Pare & Redis, 2013), skepticism towards environmental claims (Mohr, Eroǧlu, & Ellen, 1998), media skepticism (Tsfati, 2003) or psychiatric skepticism

(Swami & Furhnam, 2011). There has also been some interest in epistemic vigilance - skepticism in broader communication context (Mercier, 2017; Sperber et al.,2010). However, there is no work on, or measure of general, dispositional scepticism, with items that are context-free. The skepticism subscale may serve as a stepping stone for research on dispositional skepticism as it offers context free items that tap into the tentative doubt that characterizes this construct.

Apart from constructing a psychometrically sound measure of the general tendency to believe in conspiracy theories, the results from our study also contribute towards conceptual clarity regarding the construct of conspiracy theories. Some scholars (e.g. Pigden, 2006) have suggested that real historical conspiracies are not psychologically distinct from conspiracy theories, but others reserve the latter term for claims that are normatively considered to be irrational and implausible (Bjerg & Presskorn-Thygesen, 2017). The current research is more consistent with the second view; participants were not equally prone to believe in all kind of conspiracies. While they believed behaviors such as soliciting kickbacks are likely, they rated the scenarios pretest participants termed “conspiracy theories” to be unlikely. Furthermore, the intercorrelations among ratings of conspiracy theories were very high, just like the

intercorrelations among the corruption cases, but correlations between conspiracy theories and corruption cases were small to nonexistent, and factor analysis suggested that these two sets of claims cluster on different dimensions.

While some psychologists have had negative view of conspiracy theories (e.g. Swami et al. 2014), authors from other fields, have tended to take a more sympathetic view. For example,

Fenster (1999) argues that conspiracy theories unfairly tend to be delegitimized as a form of political or intellectual disagreement. Others suggests that they should be seen as a form of legitimate skepticism arising as a result of the increasingly secretive world or knowledge about real life conspiracies, and thus are not unfounded (e.g. Olmsted, 2009). Yet other authors suggest that conspiracy theories might encourage a useful, critical attitude towards authorities and promote transparency (e.g. Basham, 2003; Husting & Orr, 2007); some even consider them vital to a functioning democracy (Bjerg & Presskorn-Thygesen, 2017). Our findings indicate that psychologists and cultural studies scholars may be talking about different phenomena, which perhaps map on to our two dimensions of conspiracy mentality, with skepticism underlying the healthy suspicion which some scholars applaud, of more mundane abuses of power. Indeed,

Byford (2011) remarks that what some scholars include under the category "conspiracy theory" differs from its conventional meaning. At the very least, our results suggest that “conspiracy theory” as a psychological construct encompasses only a subset of the beliefs that tend to be labeled as conspiracy beliefs by the academic community.

It should be noted however, that while we interpret the two dimensions of our scale in terms of “plausible” and “implausible” claims, plausibility is neither assessed nor manipulated in our work; that distinction is based on the claims’ a priori subjective likelihood. For example, the claim that the government was involved in the 9/11 terror attacks is distinct from the claim that a

CEO bribed a UN official, not because the former is less likely to be true, but because people are less likely to believe it to be true. We chose claims that, on their face, differ greatly in subjective plausibility, but there will of course be other claims that are more ambiguous, as well as claims that vary in their plausibility over time. Our hypothesis, ripe for test in future research, is that the predictiveness of conspiracy ideation and skepticism will track the subjective plausibility between and within claims.

The current work has limitations, of course. Most obviously, our studies relied largely

(but not exclusively) on American on MTurk participants, and one might question the generalizability of the results despite our demonstration of metric and structural equivalence in two other countries. Indeed, some authors have stressed the unique place of conspiracy theories in American culture (e.g. Knight, 2000). However, other authors stress the unique place of conspiracy theories in the Middle East (e.g. Pipes, 1996), and in Russia (Ortmann &

Heathershaw, 2012), and it is an empirical question whether such theories are “unique” to any subset of cultures. Given the fundamental motivations proposed to underlie conspiracy mentality, however, and absent evidence to the contrary, we suggest that conspiracy ideation is a cultural universal. However, cross-cultural work on belief in conspiracy theories is sorely lacking, and we hope that the current measure – designed to be generic enough to be applicable in any context

– will provide a tool for researchers to test for similarities and differences across cultures.

Electronic Supplementary Material

The electronic supplementary material is available with the online version of the article at https://econtent.hogrefe.com/doi/suppl/10.1027/1864-9335/a000381

ESM 1: Conspiracy theory definitions.sav

This file contains the 210 definitions of conspiracy theories used to develop the initial item pool.

ESM 2: Predictive Validity Items.doc

This file lists all the items that were used in Study 2.

ESM 3: Standardized regression coefficients.doc

This file contains the standardized regression coefficients obtained in Study 2 and Study 4.

References:

Abalakina-Paap, M., Stephan, W.G., Craig, T., & Gregory, W.L. (1999). Beliefs in

conspiracies. Political Psychology, 20 (3), 637-647. https://doi.org/10.1111/0162-

895X.00160

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.

https://doi.org/10.1080/00313220601118751

Barkun, M. (2003). A Culture of Conspiracy: Apocalyptic Visions in Contemporary America.

Berkeley, CA: University of California Press.

https://doi.org/10.1525/california/9780520238053.001.0001

Basham, L. (2003). Malevolent global conspiracies. Journal of Social Philosophy, 34 (1), 91-

103. https://doi.org/10.1111/1467-9833.00167.

Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate: a practical and

powerful approach to multiple testing. Journal of the Royal Statistical Society, 57, 289-

300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x

Bjerg, O., & Presskorn-Thygesen, T. (2017). Conspiracy theory: Truth Claim or Language

Game. Theory Culture & Society, 34, 137-159. https://doi.org/10.1177/0263276416657880

Brotherton, R. (2013). Towards a definition of “conspiracy theory”. PsyPAG Quarterly, 88 (3),

9–14.

Brotherton, R. (2015). Suspicious minds: why we believe conspiracy theories. New York:

Bloomsbury Sigma.

Brotherton, R., & Eser, S. (2015). Bored to fears: Boredom proneness, paranoia, and conspiracy

theories. Personality and Individual Differences, 80, 1–5.

https://doi.org/10.1016/j.paid.2015.02.011.

Brotherton, R., French, C., & Pickering, A. (2013). Measuring Belief in Conspiracy Theories:

The Generic Conspiracist Beliefs Scale. Frontiers in Psychology, 4:279.

https://doi.org/10.3389/fpsyg.2013.00279.

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. https://doi.org/10.3389/fpsyg.2013.00225

Byford, J. (2011). Conspiracy theories: A critical introduction. Basingstoke UK: Palgrave

MacMillan. https://doi.org/10.1057/9780230349216

Byrne, B. M. (2009). Structural equation modeling with AMOS : basic concepts, applications,

and programming, second edition. Retrieved from https://ebookcentral-proquest-com.

Cai, L., Maydeu-Olivares, A., Coffman, D.L., & Thissen, D. (2006). Limited information

goodness-of-fit testing of item response theory models for sparse 2p tables. British

Journal of Mathematical and Statistical Psychology, 59, 173-194.

https://doi.org/10.1348/000711005X66419

Cai, L., Thissen, D., & du Toit, S. H. C. (2017). IRTPRO 4.1 for Windows [Computer

software]. Skokie, IL: Scientific Software International, Inc.

Cichocka, A., Marchlewska, M., Golec de Zavala, A., & Olechowski, M. (2016). ‘They will not

control us’: Ingroup positivity and belief in intergroup conspiracies. British Journal of

Psychology, 107, 556–576. https://doi.org/10.1111/bjop.12158

Cichocka, A., Marchlewska, M., Golec de Zavala, A., & Olechowski, M. (2015). Grandiose

delusions. Collective narcissism, secure in-group identification, and belief in conspiracies.

In Bilewicz, Cichochka & Soral (Eds.) The Psychology of Conspiracy. London: Routledge.

https://doi.org/10.4324/9781315746838

Chen, F. (2007). Sensitivity of Goodness of Fit Indexes to Lack of Measurement

Invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14, 464-504.

https://doi.org/10.1080/10705510701301834

Cheung, G.W., & Rensvold, R.B. (2002). Evaluating Goodness-of-Fit Indexes for Testing

Measurement Invariance. Structural Equation Modeling: A Multidisciplinary

Journal, 9, 233-255, https://doi.org/10.1207/S15328007SEM0902_5

Coady, D. (2007a). Conspiracy Theories. Episteme: A Journal of Social Epistemology, 42, 131-

134. https://doi.org/10.1353/epi.2007.0019

Coady, D. (2007b). Are Conspiracy Theorists Irrational? Episteme: A Journal of Social

Epistemology, 42, 193-204. https://doi.org/10.3366/epi.2007.4.2.193

Conte, H. R., Weiner, M. B., & Plutchik, R. (1982). Measuring death anxiety: Conceptual,

psychometric, and factor-analytic aspects. Journal of Personality and Social

Psychology, 43(4), 775–785. https://doi.org/10.1037/0022-3514.43.4.775

Cortina, J. (1993). What is coefficient alpha: an examination of theory and applications. Journal

of Applied Psychology, 78, 98-104. https://doi.org/10.1037/0021-9010.78.1.98

Cubitt, G.(1989). Conspiracy myths and conspiracy theories. Journal of the Anthropological

Society of Oxford, 20, 12-26.

Cubitt, G. (1993). The Jesuit myth: conspiracy theory and politics in nineteenth-century France.

Oxford: Clarendon Press.

Dagnall, N., Denovan, A., Drinkwater, K., Parker, A., & Clough, P. (2017). Statistical Bias and

Endorsement of Conspiracy Theories. Applied Cognitive Psychology, 31, 368-378.

https://doi.org/10.1002/acp.3331

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 (8), 1289–1293. https://doi.org/10.1016/j.paid.2011.02.027

De Haven-Smith, L. (2010). Beyond Conspiracy Theory: Patterns of High Crime in American

Government. American Behavioral Scientist, 53, 795-825.

https://doi.org/10.1177/0002764209353274

De-Haven Smith, L. & Witt, M.T. (2012). Conspiracy Theory Reconsidered Responding to

Mass Suspicions of Political Criminality in High Office. Administration and Society, 45,

267- 295. https://doi.org/10.1177/0095399712459727

Demerens, F., Pare, J-L., & Redis, J. (2013). Investor scepticism and creative accounting: the

case of a French SME listed on Alternext. International Journal of Business, 18, 59-80.

Dentith, M. (2014). The philosophy of conspiracy theories. New York, NY: Palgrave Macmillan.

Drinkwater, K., Dagnall, N., Parker, A. (2012). Reality Testing, Conspiracy Theory, and

Paranormal Beliefs. The Journal of Parapsychology, 76(1), 57-77.

Duran, N.D., Nicholson, S.P., & Dale, R. (in press). The hidden appeal and aversion to political

conspiracies as revealed in the response dynamics of partisans. Journal of Experimental

Social Psychology. https://doi.org/10.1016/j.jesp.2017.07.008

Ender, P. (n.d). FAPARA.HLP.HTM. Retrieved from

https://stats.idre.ucla.edu/stata/ado/analysis/fapara-hlp-htm/

Fekete, L. (2011). The Muslim Conspiracy Theory and the Oslo Massacre. Race and Class, 53,

30-47.

Internal Revenue Service. (2017a). Examples of Public Corruption Investigations for the fiscal

year 2016. Retrieved from: https://www.irs.gov/compliance/criminal-

investigation/examples-of-public-corruption-investigations-fiscal-year-2016

Internal Revenue Service. (2017b). Examples of Public Corruption Investigations for the fiscal

year 2015. (2017, September 29). Retrieved from:

https://www.irs.gov/compliance/criminal-investigation/examples-of-public-corruption-

investigations-fiscal-year-2015

Feningstein, A., & Vanable, P. A. (1992). Paranoia and self-consciousness. Journal of

Personality and Social Psychology, 62(1), 129-138. https://doi.org/10.1037/0022-

3514.62.1.129.

Fenster, M. (1999). Conspiracy Theories: Secrecy and Power in American culture. Minneapolis:

University of Minnesota Press.

Field, A. (2009). Discovering Statistics Using SPSS. London: SAGE Publications Ltd.

Ford, G.T., Smith, D.B., & Swasy, J.L (1990). Consumer scepticism of advertising claims:

testing hypotheses from economics od information. Journal of Consumer Research, 16,

433-441. https://doi.org/10.1086/209228

Forehand, M.R. & Grier, S. (2003). When is honesty the best policy? The effect of stated

company intent on consumer scepticism. Journal of Consumer Psychology, 13, 349-356.

https://doi.org/10.1207/S15327663JCP1303_15

Galliford, N. & Furnham, A. (2017). Individual difference factors and beliefs in medical and

political conspiracy theories. Scandinavian Journal of Psychology 58, 422–428.

https://doi.org/10.1111/sjop.12382

Goertzel, T. (1994). Belief in Conspiracy Theories. Political Psychology,15(4), 731-742.

https://doi.org/10.2307/3791630

Gosling, S. D., Rentfrow, P. J., & Swann, W. B., Jr. (2003). A Very Brief Measure of the Big

Five Personality Domains. Journal of Research in Personality, 37, 504-528.

https://doi.org/10.1016/S0092-6566(03)00046-1

Green, R., & Douglas, K. (2018). Anxious attachment and belief in conspiracy theories.

Personality and Individual Differences, 125, 30-37.

https://doi.org/10.1016/j.paid.2017.12.023

Hagen, K. (2017). Conspiracy Theories and the Paranoid Style: Do Conspiracy Theories Posit

Implausibly Vast and Evil Conspiracies? Social Epistemology, 32(1), 24-40.

https://doi.org/10.1080/02691728.2017.1352625

Hofstadter, R. (1966). The paranoid style in American politics. In R. Hofstadter (Ed.), The

paranoid style in American politics and other essays (pp. 3 – 40). New York, NY: Knopf.

Horn, J.L. (1965). A rationale and test for the number of factors in factor analysis.

Psychometrika, 30(2), 179-185. https://doi.org/10.1007/BF02289447

Hurtt, R.K. (2010). Development of a scale to measure professional scepticism. Auditing: A

Journal of Practice & Theory, 29, 149-171. doi:10.2308/aud.2010.29.1.149

Husting, G., & Orr, M. (2007). Dangerous machinery: "conspiracy theorist" as a transpersonal

strategy of exclusion. Symbolic Interaction, 30(2), 127-150.

https://doi.org/10.1525/si.2007.30.2.127

Imhoff, R. & Bruder, M. (2014). Speaking (un-)truth to power: Conspiracy Mentality as a

generalized political attitude. European Journal of Personality, 28(1), 25-43.

https://doi.org/10.1002/per.1930

Imhoff, R., & Lamberty, P. K. (2017). Too special to be duped: Need for uniqueness motivates

conspiracy beliefs. European Journal of Social Psychology, 47, 724–734.

https://doi.org/10.1002/ejsp.2265.

Imhoff, R. and Lamberty, P. (in press). How paranoid are conspiracy believers? Toward a more

fine‐grained understanding of the connect and disconnect between paranoia and belief in

conspiracy theories. European Journal of Social Psychology.

https://doi.org/10.1002/ejsp.2494

Jane, E., & Fleming, C. (2014). Modern Conspiracy: The Importance of Being Paranoid. New

York: Bloomsbury Publishing.

Jensen, T. (2013). Democrats and Republicans differ on conspiracy theory beliefs. Retrieved

from https://www.publicpolicypolling.com/polls/democrats-and-republicans-differ-on-

conspiracy-theory-beliefs/ on 2 Nov 2017

Jensen, T. (2016). GOP Quickly Unifies Around Trump; Clinton Still Has Modest Lead.

Retrieved from https://www.publicpolicypolling.com/polls/gop-quickly-unifies-around-

trump-clinton-still-has-modest-lead/ on 02 Nov 2017.

Jolley, D., & Douglas, K.M. (2014a). The effects of anti-vaccine conspiracy theories on

vaccination intentions. PLOS ONE, 9(2): e89177.

https://doi.org/10.1371/journal.pone.0089177;

Jolley, D., & Douglas, K.M. (2014b). The social consequences of conspiracism: Exposure to

conspiracy theories decreases the intention to engage in politics and to reduce one’s carbon

footprint. British Journal of Psychology, 105, 35-56. https://doi.org/10.1111/bjop.12018

Knight, P. (2000). Conspiracy culture: From Kennedy to the X – files. London: Routledge.

Kreko, P. (2015). Conspiracy theory as collective motivated cognition. In M. Bilewicz, A.

Cichochka & W. Soral (Eds.) The Psychology of Conspiracy Theories. London: Routledge.

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), 1–14. https://doi.org/10.5334/irsp.8

Lantian, A., Muller, D., Nurra, C., & Douglas, K. M. (2017). “I know things they don’t know!”

the role of need for uniqueness in belief in conspiracy theories. Social Psychology, 48, 160-

173. https://doi.org/10.1027/1864-9335/a000306

Lantian, A., Muller, D., Nurra, C., Klein, O., Berjot, S. and Pantazi, M. (in press). Stigmatized

beliefs: Conspiracy theories, anticipated negative evaluation of the self, and fear of social

exclusion. European Journal of Social Psychology. https://doi.org/10.1002/ejsp.2498

Leone, L., Giacomantonio, M & Lauriola, M. (2017). Moral foundations, worldviews, moral

absolutism and belief in conspiracy theories. International Journal of Psychology.

https://doi.org/10.1002/ijop.12459

Leone, L., Giacomantonio, M., Williams, R., & Michetti, D. (2018). Avoidant attachment style

and conspiracy theories. Personality and Individual Differences, 134, 329-336.

https://doi.org/10.1016/j.paid.2018.06.043

Lobato, E., Mendoza, J., Sims, V., & Chin, M. (2014). Examining the relationship between

conspiracy theories, paranormal beliefs, and pseudoscience acceptance among a university

population. Applied Cognitive Psychology, 28, 617-625. https://doi.org/10.1002/acp.3042

Mashiri, A. & Zaduqisti, E. (2014). We believe in your conspiracy if we distrust you: The role of

intergroup distrust in structuring the effect of Islamic identification, competitive

victimhood, and group incompatibility on belief in conspiracy theory. Journal of Tropical

Psychology, 4,E11. https://doi.org/10.1017/jtp.2014.11

Mashuri, A. & Zaduqisti, E. (2015). The effect of intergroup threat and social identity salience

on the belief in conspiracy theories over terrorism in Indonesia: Collective angst as

mediator. International Journal of Psychological Research, 8, 24-35.

https://doi.org/10.21500/20112084.642

Maydeu-Olivares, A. & Joe, H. (2006). Limited information goodness-of-fit testing in

multidimensional contingency tables. Psychometrika, 71, 713-732.

https://doi.org/10.1007/s11336-005-1295-9

Maydeu-Olivares, A., & Joe, H. (2005). Limited and full information estimation and testing in 2n

contingency tables: A unified framework. Journal of the American Statistical

Association, 100, 1009–1020. https://doi.org/10.1198/016214504000002069

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. https://doi.org/10.1037/0022-3514.37.5.637

Mercier, H. (2017). How gullible are we? A review of the evidence from psychology and social

science. Review of General Psychology, 21(2), 103-122.

http://dx.doi.org/10.1037/gpr0000111

Mohr, L. A., Eroǧlu, D. and Ellen, P. S. (1998). The Development and Testing of a Measure of

Skepticism Toward Environmental Claims in Marketers' Communications. Journal of

Consumer Affairs, 32, 30-55. https://doi.org/10.1111/j.1745-6606.1998.tb00399.x

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(8), 1007-1011. http://dx.doi.org/10.1016/j.paid.2011.08.011

Obermiller, C., & Spangenberg, E.R. (1998). Development of a Scale to Measure Consumer

Skepticism Toward Advertising. Journal of Consumer Psychology, 7(2), 159-186.

https://doi.org/ 10.1207/s15327663jcp0702_03

Oliver, E.J. & Wood, T.J. (2014). Conspiracy Theories and the Paranoid Style(s) of Mass

Opinion. American Journal of Political Science, 58(4), 952-966.

Olmsted, K. (2009). Real Enemies: Conspiracy Theories and American Democracy, World War I

to 9/11. New York: Oxford University Press.

Orlando, M., & Thissen, D. (2000). Likelihood-based item fit indices for dichotomous item

response theory models. Applied Psychological Measurement, 24, 50-64.

https://doi.org/10.1177/01466216000241003

Orlando, M. & Thissen, D. (2003). Further investigation of the performance of S- χ2: An item fit

index for use with dichotomous item response theory models. Applied Psychological

Measurement, 27, 289-298. https://doi.org/10.1177/0146621603027004004

Orosz, G., Krekó, P., Paskuj, B., Tóth-Király, I., Bo the̋ , B., & Roland-Lévy, C. (2016). Changing

Conspiracy Beliefs through Rationality and Ridiculing. Froniers in Psychology, 7:1525.

https://doi.org/10.3389/fpsyg.2016.01525

Ortmann, S. & Heathershaw, J. (2012). Conspiracy Theories in the Post‐Soviet Space. The

Russian Review, 71, 551-564. https://doi.org/10.1111/j.1467-9434.2012.00668.x

Pacini, R., & Epstein, S. (1999). The relation of rational and experiential information processing

styles to personality, basic beliefs, and the ratio-bias phenomena. Journal of Personality

and Social Psychology, 76(6), 972-987. https://doi.org/10.1037/0022-3514.76.6.972

Peirce, J. W. (2007). PsychoPy-Psychophysics software in Python. Journal of Neuroscience

Methods, 162, 8-13. https://doi.org/ 10.1016/j.jneumeth.2006.11.017

Pigden, C. (2006). Complots of Mischief. In D. Coady (ed.), Conspiracy Theories: The

Philosophical Debate, pp. 139-166. Aldershot: Ashgate.

Pipes, D. (1996). The hidden hand: Middle East fears of conspiracy. New York : St. Martin's

Press

Pipes, D. (1997). How the paranoid style flourishes and where it comes form. New York: Free

Press.

Robins, R., and Post, J. (1997). Political Paranoia: The Psychopolitics of Hatred. New Haven,

CT: Yale University Press.

Samejima, F. (1969). Estimation of a latent ability using a response pattern of graded scores.

Psychometrika Monograph Supplement, 34. https://doi.org/10.1002/j.2333-

8504.1968.tb00153.x

Schmitt, N. (1996). Uses and abuses of coefficient alpha. Psychological Assessment, 8, 350-353.

https://doi.org/10.1037/1040-3590.8.4.350

Sijtsma, K. (2009). On the use, the misuse, and the very limited usefulness of Cronbach’s Alpha.

Psychometrika, 74, 107-120. https://doi.org/10.1007/s11336-008-9101-0

Sperber, D., Clement, F., Heintz, C., Mascaro, O., Mercier, H., Origgi, G. & Wilson, D. (2010).

Epistemic vigilance. Mind & Language, 25, 359-393. https://doi.org/10.1111/j.1468-

0017.2010.01394.x

Sunstein, C. R. & Vermeule, A. (2009). Conspiracy theories: Causes and cures. Journal of

Political Philosophy, 17, 202–227. https://doi.org/10.1111/j.1467-9760.2008.00325.x

Swami, V. & Furnham, A. (2011). Preliminary examination of the psychometric properties of the

Psychiatric Scepticism Scale. Scandinavian Journal of Psychology, 52, 399-

403. https://doi.org/ 10.1111/j.1467-9450.2011.00881.x

Swami, V. & Furnham, A. (2012). Examining conspiracist beliefs about the disappearance of

Amelia Earhart. Journal of General Psychology, 139, 244–259.

https://doi.org/10.1080/00221309.2012.697932

Swami, V., Barron, D., Weis, L., Voracek, M., Stieger, S., & Furnham, A. (2017). An

examination of the factorial and convergent validity of four measures of conspiracist

ideation, with recommendations for researchers. PLOS ONE 12(2): e0172617.

https://doi.org/10.1371/journal.pone.0172617

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.

https://doi.org/10.1002/acp.1583

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.

https://doi.org/10.1111/j.2044-8295.2010.02004.x

Swami, V., Voracek, M., Stieger, S., Tran, U., & Furnham, A. (2014). Analytic thinking reduces

beliefs in conspiracy theories. Cognition, 133(3), 572-585.

https://doi.org/10.1016/j.cognition.2014.08.006

Swami, V., Pietschnig, J., Tran, U. S., Nader, I., Stieger, S. and Voracek, M. (2013). Lunar Lies:

the impact of informational framing and individual differences in shaping conspiracist

beliefs about the moon landings. Applied Cognitive Psychology 27: 71–

80. https://doi.org/10.1002/acp.2873

Swami, V., Weis, L., Lay, A., Barron, D. & Furnham, A. (2016). Associations between belief in

conspiracy theories and the maladaptive personality traits of the personality inventory for

DSM-5. Psychiatry Research, 236, 86-90. https://doi.org/10.1016/j.psychres.2015.12.027

Tobacyk, J. (2004). A Revised Paranormal Belief Scale. The International Journal of

Transpersonal Studies, 23, 94-98. https://doi.org/10.1037/t14015-000

Tsfati, Y. (2003). Does Audience Skepticism of the Media Matter in Agenda Setting? Journal of

Broadcasting & Electronic Media, 47, 157-176.

https://doi.org/10.1207/s15506878jobem4702_1

Uenal, F. (2016). The “Secret Islamization” of Europe: Exploring Integrated Threat Theory for

Predicting Islamophobic Conspiracy Stereotypes. International Journal of Conflict and

Violence, 10, 94-108. https://doi.org/10.4119/UNIBI/ijcv.499

Uscinski, J. E., & Parent, J. M. (2014). American conspiracy theories. New York: Oxford

University Press. https://doi.org/10.1093/acprof:oso/9780199351800.001.0001 van Prooijen, J.-W., & van Lange, P. A. M. (2014). The social dimension of belief in

conspiracy theories. Chapter in J.-W. van Prooijen and 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.

https://doi.org/10.1017/CBO9781139565417 van Prooijen, J-W., & Acker, M. (2015). The influence of control on belief in conspiracy

theories: Conceptual and applied extensions. Applied Cognitive Psychology, 29, 753-761.

https://doi.org/10.1002/acp.3161

Wilson, M. W. & Rose, C. (2014). The role of paranoia in a dual-process motivational model of

conspiracy belief. In van Prooijen & van Lange (Eds.), Power Politics and Paranoia:

Why People are Suspicious of Their Leaders (273-291). Cambridge: Cambridge

University Press. https://doi.org/10.1017/CBO9781139565417

Wong, C.-S., & Law, K.S. (2002). The effects of leader and follower emotional intelligence on

performance and attitude: An exploratory study. The Leadership Quarterly, 13(3), 243-

274. https://doi.org/10.1016/S1048-9843(02)00099-1

Wood, M. J. (2016). Some Dare Call It Conspiracy: Labeling Something a Conspiracy Theory

Does Not Reduce Belief in It. Political Psychology, 37, 695-705.

https://doi.org/10.1111/pops.12285

Wood, M. J. (2017). Conspiracy suspicions as a proxy for beliefs in conspiracy theories:

Implications for theory and measurement. British Journal of Psychology, 108: 507–527.

https://doi.org/10.1111/bjop.12231

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.

https://doi.org/10.1177/1948550611434786.

Zhang, X.J., Ko, M., & Carpenter, D. (2016). Development of a scale to measure skepticism

toward electronic word-of-mouth. Computers in Human Behavior, 56, 198-208.

https://doi.org/10.1016/j.chb.2015.11.042

Chapter 3: Does Perceived Lack of Control Lead to Conspiracy

Theory Beliefs?

Does Perceived Lack of Control Lead to Conspiracy Theory Beliefs?

Ana Stojanov

Department of Psychology

University of Otago

Jesse M. Bering

Centre for Science Communication

University of Otago

Jamin Halberstadt

Department of Psychology

University of Otago

Address for correspondence:

Ana Stojanov

William James Building

275 Leith Walk, Dunedin 9016, New Zealand e-mail: [email protected]

Word count: 6763

Abstract

It is widely believed that conspiracy theory beliefs are the product of perceived lack of control. However, to date there is mixed evidence, at best, to support this claim. We consider the reasons why conspiracy theory beliefs do not appear to be based in any straightforward way on control beliefs, interrogating existing findings and presenting new data that call the relationship into question. Across six studies, we observed no effect of control manipulations on conspiracy theory beliefs, while replicating previously reported correlational evidence of their association.

The results suggest that conspiracy beliefs are not suitable for compensating for threats to control, although the discrepancy between experimental and correlational effects opens the possibility that the causal link may run in the opposite direction.

Keywords: conspiracy theories, compensatory control, lack of control, control threat,

Hurricane Katrina, powerlessness, meaning making, sense making

The world is ruled by a secret cabal. The US government was involved in the 9/11 attacks. HIV is a man-made virus. Such statements, which are promoted on the internet and elsewhere, are examples of conspiracy theories: implausible, unwarranted claims that important social events are caused by malevolent clandestine groups, that usually run in contradiction to the explanations offered by the relevant epistemic authorities, and that are embedded in a more general worldview (Stojanov & Halberstadt, 2019).

Although researchers have tried to understand conspiracy theory beliefs – which to nonbelievers appear unwarranted by evidence, if not entirely irrational – from many perspectives, including personality theory (Green & Douglas, 2018; Swami et al., 2011), intergroup identity (Cichocka, Marchlewska, Golec de Zavala & Olechowski, 2016; Mashuri &

Zaduqisti, 2014; 2015), and cognitive bias (e.g., Brotherton & French, 2014, 2015; Douglas,

Sutton, Callan, Dawtry & Harvey, 2016), arguably the most frequent accounts rely on some notion of “control.” The claim that perceived lack of control is the natural breeding ground for conspiracy theory beliefs is a recurring theme in popular science (e.g. Brotherton, 2015;

Oaklander, 2015; Pichaske & Alpert, 2018; Shermer, 2014; Yong, 2008), and many academics agree (e.g., Douglas, Sutton & Cichocka, 2017; Kossowska & Bukowski, 2015; Marchlewska,

Cichocka, & Kossowska, 2018; Wood & Douglas, 2019).

However, on closer inspection, neither the theoretical role of control in conspiratorial thinking, nor the empirical relationship between the two constructs, is entirely convincing.

Although specific accounts differ in their emphasis, control accounts of conspiracy theories generally assume that the perceived loss of personal control is threatening to one’s need for order

(e.g., Kay, Whitson, Gaucher, & Galinsky, 2009) or one’s schemas about the world (e.g., Heine,

100

Proulx, & Vohs, 2006). Conspiracy beliefs help compensate for these threats by ascribing control to external entities or organizations (e.g. Kay & Eibach, 2013; Kay, Sullivan & Landau, 2015;

Rutjens & Kay, 2016), or by creating new meaning frameworks via novel connections among unrelated events (van Prooijen, in press). For example, perceiving a connection between a tornado and the unusually long contrails observed the previous day, provides a basis for believing that the government (or some other entity) has deliberately sprayed chemicals in the sky to cause the tornado.

However, it seems equally plausible that conspiracy theories would have the opposite effect on the perception of control and order. A common theme among conspiracy theories is that reality is not what it seems; they are counter-normative narratives that suggest institutions and explanations that we take for granted are illusions – a revelation that, if taken seriously, seems bound to challenge the larger meaning-making systems individuals have come to rely on.

Furthermore, although people desire order, they seek to optimize, not to maximize it, and it could be argued that conspiracy theories provide “too much” order, structure, and meaning by accounting for errant data, connecting seemingly unconnected events and leaving nothing unexplained (Clarke, 2002; Keeley, 1999). Finally, even if conspiracy theories represent an effective means of restoring order and structure, they are not the only means (Kay,

Gaucher, Mcgregor & Nash, 2010; Kay, Gaucher, Napier, Callan, & Laurin, 2008; Kay,

Whitson, Gaucher & Galinsky, 2009; Landau, Kay & Whitson, 2015), and, it would seem, a means of last resort. While control accounts do not require that sources of control or meaning be benevolent, people theoretically favour control systems that are both “culturally accessible” and

“socially acceptable” (Kay et al., 2008; Kay & Sullivan, 2013). In most—if not all—conspiracy theories, however, the alleged conspirators are malevolent (van Prooijen, 2018), thus making

101

them poor candidates for compensatory processes when numerous effective and socially sanctioned alternatives exist.

The experimental evidence for conspiracies as compensatory sources of personal control is as ambivalent as its theoretical rationale. In a typical experiment, researchers ask participants to write about a time they did not have control, and to compare these participants’ conspiracy beliefs with a control group that is asked to describe an event when they felt in complete control

(or about a neutral event). Some studies report stronger conspiracy beliefs in the low/neutral control group compared to the high control group (e.g. Whitson & Galinsky, 2008; van Prooijen

& Acker, 2015), but others report weaker ones (e.g. van Elk & Lodder, 2018), and yet others

(e.g. Hart & Graether, 2018; Nyhan & Zeitzoff, 2018) report no effects of control on conspiracy beliefs at all.

One factor in the variability of results is the corresponding variability – and sometimes dubious validity – of how belief in conspiracy theories is operationalized and measured. For example, Whitson and Galinsky (2008, Study 4) found that participants recalling uncontrollable situations were more likely to perceive patterns in ambiguous stimuli (e.g., images in visual noise) and links between events (e.g., an increase in e-mail correspondence and a lack of a promotion). However, while conspiracy theories are related to illusory pattern perception (van

Prooijen, Douglas & de Inocencio, 2018), they are not reducible to it. And while novel connections between seemingly unrelated events can, of course, be the building blocks of conspiracy theories, such theories tell a more complex story that taps into a broader worldview.

For example, seeing a connection between the “umbrella man” (a man who opened and lifted an umbrella when Kennedy’s limousine approached) and the assassination of Kennedy is a starting

102

point for building a conspiracy theory, but it is not a conspiracy theory per se (e.g., that

Kennedy’s assignation was an organized Soviet-backed plot carried out for political purposes).

In another operationalization of conspiracy belief, Van Prooijen and Acker (2015) asked participants about the construction of the North-South metro line in Amsterdam, by posing situations that were technically conspiracies, but also plausible abuses of power warranting legitimate skepticism. Accusations such as, “The city council transferred parts of the budget to the bank accounts of others”, and “Members of the city council received money from construction companies to set this plan in motion,” describe political corruption, which differ from “conspiracy theories” in scope, plausibility, official endorsement, and psychological interest. Indeed, recent evidence indicates that belief in conspiracy theories and belief in corruption are two distinct psychological constructs, predicted by different aspects of a larger construct (“conspiracy mentality”; Stojanov & Halberstadt, 2019).

Even studies that explicitly ask about beliefs in specific conspiracy theories may not be well suited for assessing the construct of conspiracy theory belief. For one, participants can easily recognize such items as “conspiracy theories,” whose negative connotations are likely to produce socially desirable responses (Stojanov & Halberstadt, 2019). Moreover, using specific conspiracy beliefs limits the generalizability of the findings to very particular and culturally specific claims, and may say little about a person’s general tendency to employ conspiratorial logic. More relevant, we argue, is an individual’s tendency to engage in conspiracy thinking, rather than their belief in any particular claim.

Given the significance and potential consequences of widespread conspiracy beliefs

(Douglas & Sutton, 2018; Douglas, Sutton, Jolley & Wood, 2015), and the plausible but largely unsubstantiated role of control in their appeal, we here report three studies (see Table 3) to test

103

the effects of lack of control on conspiracy theory beliefs using a standard priming paradigm and a validated measure of conspiracy ideation, which reflects the belief that a powerful entity lies behind significant social or political events and that the conventional (official) truth is not the

“real” truth. A fourth study confirms the results with a measure of specific claims, and two final studies examine two of the most likely moderators of control effects: the degree of control that a particular conspiracy claim affords; and the effectiveness of alternative means of restoring control. Although we conceptually replicate previous correlational work – lower control beliefs were associated with stronger conspiracy ideation – we find no evidence for a causal effect.

Study 1

Method

Participants. In this and subsequent studies, sample size was determined a priori.

Detecting an effect of f = 0.2, the average effect size reported in psychology research (Richard,

Bond & Stokes-Zoota, 2003) with alpha = 0.05 and power of 80%, required 246 participants.

Expecting some attrition, we recruited 301 Mechanical Turk (MTurk) workers (105 male, 195 female, 1 other), who completed an online questionnaire designed and presented in the Qualtrics survey environment. The mean age of the sample was M = 36.20 (range 18-79 years, SD =

12.24). The majority had Bachelor’s degree (41.2%), while 27.9% had a high school degree,

16.6% Master’s degree, 3% Doctoral degree, 2.7% no degree; 8.6% reported their education level as “other”.

Materials and Procedure. After providing informed consent, participants were informed that, as part of a study on memory, they would be asked to recall and describe an event twice,

104

once at the beginning of the study and once “after a break interval,” during which they completed the key dependent variables. Following Whitson and Galinsky (2008), we used the to- be-recalled event as our manipulation of control. The low control group described an event in which they “did not have any control over a situation,” while the high control group described an event in which they “were in complete control of the situation.” Both groups were instructed to include “what happened, how you felt, etc.,” and to write at least six lines of text. A neutral group was asked to recall and write about their dinner the previous night. The wording of the low and high control manipulation was taken verbatim from Whitson & Galinsky (2008) with the exception of the instruction for the length of the text.

To assess the effectiveness of the manipulation we measured participants’ feelings of control on Pearlin and Schooler’s (1978), Mastery Scale. It consists of seven items ( e.g., I have little control over the things that happen to me) that capture the extent to which people see themselves as being in control over their life (Pearlin, Menaghan, Lieverman & Mullan, 1981), and which has been used in numerous studies as a measure of perceived control (e.g. Goode,

Keefer, Branscombe & Molina, 2017; Ma & Kay, 2017). The scale was anchored at 1 (strongly disagree) and 4 (strongly agree). Five items were reverse scored, such that higher scores indicate higher levels of control. In the current study Cronbach’s alpha was 0.84.

Conspiracy beliefs were measured on the conspiracy theory ideation subscale of Stojanov and Halberstadt’s (2019) Conspiracy Mentality Scale. The subscale consists of seven items capturing the abstract premises of conspiratorial thinking, such as the belief in a powerful entity that controls significant social or political events. (A second dimension of the scale, measuring rational scepticism, is not relevant to the current study.) The subscale has good psychometric characteristics (in the current sample Cronbach’s alpha was 0.91) and convergent, divergent and

105

predictive validity, and its construct validity has been confirmed in United States, New Zealand and North Macedonia populations (Stojanov & Halberstadt, 2019). Participants rated their agreement with each statement on a 1-to-7 scale anchored at “strongly disagree” and “strongly agree.” Each subscale was presented in a separate block, always in the same order (with the dimension of interest, conspiracy theory ideation, presented first). An attention check item (“To make sure you read attentively please select strongly agree”), was included in the second block.

After completing the dependent measure, participants were asked, in line with the cover story, to describe their event a second time (memory was not analyzed). Finally, they were asked to speculate about the study’s hypotheses, to answer basic demographic questions, and to assess the quality of their data (“Did you honestly answer the question in this survey? You’ll be paid regardless of how you answer”), before being debriefed.

Results and Discussion

We report all measures, manipulations, and exclusions in this and subsequent studies.

Twenty-nine participants were excluded from the analysis: sixteen because they failed the attention check question (low control = 5, neutral = 4, high control =7); two because they assessed their data as unreliable (high control =2) and 11 ( low control = 7, high control = 4) because they expressed suspicion about the research aims3 leaving 272 participants in total4. A sensitivity analysis indicated that the study was able to detect a minimum effect size of f = 0.18.

The effectiveness of the control manipulation was tested with a one-way ANOVA on mastery scores, which revealed a significant effect of experimental condition F (2, 269) = 7.69, p

3 In studies 1-5, two research assistants coded whether the participants were naïve to the research aims. The inter- rater reliability was high (ranged from 0.8 to 0.97) and disagreements were settled by the first author. 4 Including such participants does not change the pattern of results in any of the studies reported in this paper. 106

< 0.001, f = 0.23. Post hoc Bonferroni comparisons indicated that the low control group (M =

2.64, SD = 0.50, N = 79) differed from both the neutral (M =2.96, SD =0.58, p < 0.001, N = 108) and the high control group (M =2.89, SD = 0.58, p = 0.02 , N = 85), but the high control group did not differ from the neutral group (p = 1.0 ). This is in line with other unsuccessful attempts to induce higher feelings of control compared to baseline levels (e.g. Cutright, 2012; Cutright,

Bettman, & Fitzsimons, 2013).

Contrary to the primary hypothesis, an ANOVA on conspiracy ideation indicated that the low control group (M = 3.66, SD = 1.17) had descriptively lower conspiracy beliefs than both the neutral (M = 3.99, SD = 1.27) and high control groups (M = 3.84, SD = 1.35), although the effect was not significant, F (2, 269) = 1.55, p = 0.215, f = 0.11. Excluding the high control group (in which the manipulation was ineffective), revealed a marginally significant difference t (185) = -

1.815, p = 0.07, d = 0.27; the low and high groups did not differ, t (162) = - 0.925, p = 0.36, d =

0.14.

Finally, we calculated the Pearson’s correlation coefficient between conspiracy beliefs and the mastery score, finding a significant negative relationship, r = -0.237, p < 0.0001.

In sum, despite using a standard and empirically successful manipulation of personal control (e.g. Chen, Lee, & Yap, 2017; Shepherd, Kay, Landau, & Keefer, 2011; Whitson &

Galinsky, 2008), we did not find that those whose control was threatened expressed greater conspiracy ideation; if anything, the tendency was for the opposite to be true. However, weaker feelings of control (independent of experimental condition) predicted stronger conspiracy ideation, as in previous research (e.g., Imhoff, 2015; van Elk & Lodder, 2018; van Prooijen &

Acker, 2015).

107

In Study 2, we attempted to replicate the results of Study 1 with a conceptually similar but arguably stronger manipulation of control.

Study 2

Method

Participants. Three hundred MTurk participants (137 male, 163 female) completed a questionnaire in the online survey platform Qualtrics. The mean age of the sample was 37.28 years (SD= 11.66, age range 18-71 years). The majority had an undergraduate degree (45%), a third (30.3%) had a high school degree, 14.4% had a graduate degree, 9.7% other and 0.7% no degree. None of the participants had taken part in Study 1.

Procedure and Materials. After providing informed consent, participants were randomly assigned to one of three groups. Those in the high control group were informed that we were interested in the types of situations that people experience as controllable (or in the low control group, as uncontrollable). For additional clarity, the instructions defined the construct of control, noting that “a person is ‘in control’ of something to the extent that they are able to direct or influence it if they want to.” Participants were asked to take a moment to think about some of the situations in their life where they have control (or no control), and to briefly describe ten such situations in a separate box. After providing ten examples they were administered the Mastery

Scale as a manipulation check (see Study 1). The neutral group proceeded immediately to this scale without engaging in any recall activity. As in Study 1, after completing the dependent measures, participants were invited to speculate about the experimental hypothesis before providing demographics and assessing their data quality.

Results and Discussion

108

Twenty-nine participants were excluded from analysis: nine (low control = 2, neutral = 2, high control = 3) because they failed the attention check question, one (low control) because they self-assessed their data as unreliable and 19 (low control = 4, high control = 14, neutral =1) because they were not naïve to the research aims. A sensitivity analysis indicated that the study was able to detect a minimum effect size of f = 0.19.

Next, we checked the effectiveness of the manipulation. A one-way ANOVA revealed a marginally significant effect, F (2,268) = 2.610, p = 0.075, f =0.14. A post hoc Bonferroni test showed that the low control group (M =2.67, SD = 0.54, N = 72) differed marginally (p = 0.086) from the high control group (M = 2.87, SD = 0.55, N = 83) group, but that neither of these differed from the neutral (M = 2.82, SD = 0.56, N = 116). Testing the primary hypothesis, an

ANOVA did not indicate a difference in conspiracy beliefs between condition (low control M =

3.96, SD = 1.12; neutral M = 3.77, SD = 1.29; high control M = 3.94, SD = 1.32), F(2,268) =

0.707, p=0.494, f= 0.08. As in Study 1, there was significant negative correlation between the mastery and conspiracy beliefs scores, r = - 0.217, p < 0.001.

In sum, despite successfully (albeit weakly) manipulating control and using a conceptually strong measure of conspiracy thinking, neither of two well-powered studies detected any causal evidence for the compensatory hypothesis, though both revealed a correlational relationship.

It is worth considering, however, whether the manipulation check itself might be contributing to the null findings, by providing an opportunity to restore control, and thereby eschewing participants’ need for further efforts in the form of conspiracy belief change. Hauser,

Ellsworth and Gonzalez (2018) argue that manipulation checks as simple as asking participants about their feelings of control may serves the purpose of restoring their control. We note that, if

109

this were true, it supports our contention that there are far easier ways to restore control than to endorse extreme, counter-normative belief systems. Nevertheless, we considered this possibility in Study 3. If the presence of the self-report measure of control was responsible for the null effect of the control manipulation on conspiracy ideation, then removing the manipulation check should produce the hypothesized relationship.

Study 3

Method

Participants. Two hundred and two MTurk workers (79 males, 123 females) participated. Half (51.5%) had undergraduate degree, a third (28.7%) high school degree, an eighth (12.9%) had a graduate degree, while 5.9% selected “other” as their highest level of education, and 1% reported they did not have a degree. The mean age of the sample was 39.14 years (SD= 12.62, range 20-71 years). No participants had taken part in previous studies reported in this paper.

Procedure and Materials. The procedure was identical to Study 1, except that the manipulation check was omitted from the procedure, and that control was only manipulated downward.

Results and Discussion

Seven participants were excluded from the analysis: one (neutral group) because they self-assessed their data as unreliable, four (neutral =3, low control =1) because they failed the attention check question and two (both low control group) because they were not naïve to the

110

research aims. A sensitivity analysis, indicated that the study is able to detect a minimum effect size of f = 0.2.

The low control group scored numerically higher on the conspiracy theory ideation (M =

3.33, SD = 0.96, N = 86) than the neutral group (M = 3.19, SD = 1.09, N = 109), but the difference was not significant, t (193) = 0.936, p = 0.350, f = 0.06. Thus, it does not appear that the inclusion of the manipulation check interfered with participants’ hypothesized motivation to restore control via conspiratorial thinking.

Studies 1-3, using a novel (but, we argue, more theoretically appropriate) operationalization of conspiracy beliefs, revealed no empirical support for the control hypothesis.

To explore whether similar results would be obtained with a more typical operationalization of conspiracy beliefs (i.e. as specific claims), and therefore exclude the possibility that the null effects are unique to the conspiracy ideation scale, we replicated the studies again, this time using the Beliefs in Conspiracy Theories Inventory (Swami et al., 2011).

Study 4

Method

Participants. Two hundred MTurk workers (72 male, 126 female, 2 other) participated.

A third had an undergraduate degree (39.5%), a third a high school degree (29.5%), while 19.5% had a graduate degree, 1.5% did not have a degree, and 9% had an “other” degree. The mean age of the sample was 40.85 years (SD=13.44, range 20-82 years). No participant had taken part in the previous studies reported here.

Procedure and Materials. The procedure was identical to Study 3 with the exception of the dependent measure. Instead of conspiracy ideation, we measured belief in fifteen specific

111

conspiracies (e.g., A powerful and secretive group, known as the New World Order, are planning to eventually rule the world through an autonomous world government, which would replace sovereign government), using the Belief in Conspiracy Theories Inventory (BCTI), developed and later modified by Swami and colleagues (Swami, Chamorro-Premuzic, & Furnham, 2010;

Swami et al., 2011).5 Participants rated the conspiracy theories on a 1 (completely false) to 9

(completely true) scale, with higher scores indicating higher conspiracy theory beliefs.

Results and Discussion

Eight participants (all low control group) were excluded from the analysis, one because they self-assessed their data as unreliable and seven because they were not naïve to the research aims. A sensitivity analysis indicated that the study was able to detect a minimum effect size of f

= 0.20.

The neutral group scored numerically higher on conspiracy theory beliefs (M = 3.31, SD

=1.57, N = 109) than the low control group (M = 3.27, SD =1.65, N = 83), but the difference was not significant t (190)= 0.157, p = 0.88, f = 0.01. Ostensibly, then, the null effects in Studies 1-3 do not appear to be due to the instrument used to measure conspiracy theory beliefs.

Although our four studies show little evidence for a main effect of control priming on conspiracy beliefs, it is entirely possible that additional factors moderate the effects of control -- that control motivates conspiracy beliefs in some circumstances but not others. In the final two studies, we explore two of the most plausible possibilities: the degree to which a conspiracy theory itself implies a controlling entity (Study 5); and the availability of other means of restoring control when it is threatened (Study 6).

5 We used the modified version (Swami et al, 2011). 112

Study 5

If people turn to conspiracy theories as a compensatory control mechanism, it may be that some theories are more effective at restoring perceptions of order than others. In particular, threats to control may only prompt belief in conspiracies involving a controlling entity (e.g., the theory that the U.S.S.R. started the “hippie” movement in the United States in order to undermine and weaken America’s traditional values), because the belief that someone is in control is sufficient to satisfy the higher order need for structure and meaning. Other theories in which a controlling entity is absent or only implicit (e.g., the theory that Osama bin Laden died prior to the 9/11 terrorist attacks but was used as a scapegoat) may not restore control, or do so to a lesser extent. Consistent with this proposal, Kay et al. (2008) found that participants whose control was threatened increased their belief in God, but only when God was described as controlling, rather than as a creator. Thus, in Study 5, we developed a new stimulus set to test the hypothesis that participants whose control was threatened would be inclined to vest more belief in specific conspiracy theories about controlling entities, but not in other conspiracy theories.

Method

Participants. The participants were three hundred MTurk workers (108 males, 192 females, M = 37.52, SD =12.001). The majority (44%) had an undergraduate degree, while 29% had a high school degree, 13% had a master’s degree, 2% a doctoral degree and 1% did not have a degree. The mean age was 37.54 years (range 18 to 73 years). None had taken part in our previous studies.

113

Materials. The dependent measure was developed in several stages. First, an independent group of MTurk participants (N = 210) were asked to write five conspiracy theories that they had heard before, from which we identified 112 unique exemplars. Next, two research assistants were asked to code these theories on a 9-point scale anchored at 1 (“the goal of this conspiracy is NOT control”) and 9 (“the goal of this conspiracy is control”). The intraclass correlation coefficient was r = 0.84. The final rating comprised the average of the ratings of the two research assistants. Next, a second group of 85 MTurk participants was asked to rate the conspiracy theories in terms of their prototypicality, using a seven point scale anchored at 1 (“a bad fit” with their idea of a conspiracy theory) and 9 (an “excellent fit” with their idea of a conspiracy theory). A third group of 96 MTurk participants was asked to rate the same conspiracies for likely veracity, using a nine point scale anchored at 1 (“definitely false”) and 9

(”definitely true”). The intraclass correlation coefficient for prototypicality ratings was r = 0.89 and for the veracity r = 0.96. The prototypicality and veracity scores for each conspiracy theory were obtained by averaging ratings, respectively across all participants.

These extensive pretest data were used to create seven pairs of conspiracy theories that were equated on veracity ( t (95) = 0.604, p=0.55, d= 0.06) and prototypicality (t (84) = 0.439, p

= 0.66, d= 0.05), but that each differed by at least 5 scale points in terms of control (Ms = 8.21 vs. 1.71). For the full list of the selected conspiracies, along with their control, prototypicality and veracity rating (obtained from the pre-test) see Table 1, for the full list of all pretest conspiracies see the online supplementary materials.

114

Table 1. Conspiracy theory pairs

Low control C V P M M M High control C V P M M M theory (SD) (SD) (SD) theory (SD) (SD) (SD) low neutra high low neutra high l l 1 Lyme disease is a 1.50 1.92 4.69 1.95 1.73 1.72 Soy is being used 7 1.97 4.74 1.79 1.70 2.00 cross between (1.83) (2.98) (1.57) (1.21) (1.24) to feminize men. (1.87) (2.99) (1.35) (1.19) (1.43) aids and syphilis that escaped the lab via an infected tick. 2 There is a secret 1 2.21 4.89 1.89 2.03 2.22 The USSR only 6.50 2.32 4.92 2.05 2.12 2.14 alien base on (2.04) (3.24) (1.54) (1.49) (1.70) appeared to (1.97) (2.83) (1.28) (1.34) (1.27) Antarctica collapse to cause a false sense of safety in the West. 3 Bigfoot, a human- 1.50 1.73 4.97 1.90 1.86 1.95 The weather is 9 1.86 4.99 1.78 1.72 1.85 animal hybrid, is (1.37) (2.94) (1.45) (1.33) (1.42) controlled by (1.80) (3.12) (1.46) (1.23) (1.62) a scientific multiple experiment gone governments to wrong. help various industries and hurt others. 4 There are aliens 1.50 2.81 5.07 2.34 2.39 2.61 The government 9 3.27 5.11 2.48 2.54 2.66 walking among us (2.49) (3.03) (1.72) (1.64) (1.87) planted cocaine in (2.71) (3.13) (1.88) (1.82) (2.02) right now black neighbourhoods to keep black

115

society addicted and helpless. 5 Russia sent 2 2.61 5.18 2.54 2.80 2.50 Russia started the 7.50 2.23 5.18 1.83 1.84 1.89 numerous humans (1.94) (2.63) (1.73) (1.76) (1.53) hippie movement (2.23) (3.08) (1.33) (1.33) (1.30) into space before in the US because Yuri Gagarin, but it wanted to they died there. undermine and weaken America’s strong traditional values. 6 President Obama 1.50 2.14 5.39 2.13 2.15 2.14 Jews secretly 9 2.21 5.44 1.92 1.74 2.15 is secretly a (2.33) (3.14) (1.84) (1.79) (1.80) control the (2.24) (3.15) (1.47) (1.33) (1.64) Muslim who banking system wasn’t born in and conspire for Hawaii. nefarious purposes. 7 Osama Bin Laden 3 2.27 5.28 1.92 1.92 1.89 The identification 9 2.26 5.58 1.84 1.90 2.14 dies just before (2.11) (3.12) (1.54) (1.55) (1.42) chips implanted (1.94) (2.84) (1.44) (1.44) (1.40) 9/11 but was used into household as a scapegoat. pets are being used to spy on citizens. Note: C = average control rating during pre-test, V = mean veracity rating during pre-test (and its standard deviation), P = mean prototypicality rating during pre-test (and its standard deviation). Each pair is matched on prototypicality and veracity but differs on control rating. M (SD) = the mean rating and its standard deviation for the current study.

116

Procedure. The procedure was identical to that of Study 3, except for the use of two types of specific conspiracy theories, varying in the control they imply, and the addition of a high control group as additional comparison group. Participants rated their agreement with the 14 conspiracy statements, presented in a random order, on a 7-point scale (1=strongly disagree, 7=strongly agree). Mixed among these statements was an attention check question

(i.e., “As a response to this question select strongly agree.”).

Results and Discussion

Nineteen participants were excluded from the analysis: nine (low control =3, neutral

=6) participants due to failures of attention and ten (low control = 4, high control = 6) because they were not naïve to the research aims. A sensitivity analysis (α= 0.05, 1 - β = 0.8, correlation among repeated measures = 0.8) indicated that the study is able to detect a minimum effect size of f = 0.06.

A mixed ANOVA with control manipulation (low control vs. neutral vs. high control) as between subject factor, and type of conspiracy theory (control related theory or non- control related theory) as a repeated measure, revealed no main effect of the manipulation F

(2, 278) = 0.250, p = 0.779, f =0.03, (low control: M = 2.03, SD =1.11, N = 92; neutral: M =

2.03, SD = 0.98, N = 115; high control: M = 2.13, SD = 1.12, N = 74), but a significant effect of conspiracy theory type F (1, 278) = 7.40, p = 0.007, f = 0.16, with participants expressing stronger belief in conspiracy theories that did not involve control (control related CT: M =

2.00, SD = 1.09; control non related CT: M =2.12, SD =1.14). Crucially for our purposes, there was no conspiracy theory type x control manipulation interaction F (2, 278) = 1.144, p

117

= 0.32, f =0.086. We also analysed each pair separately. The interaction was not significant for any individual pair.

In sum, we found no support for the hypothesis that control threat motivates belief in theories that hypothesize a controlling entity. In addition, participants rated the theories as more believable when they did not afford control, an unexpected result given that the two types of theories had been matched for believability in pretesting. To explore this discrepancy further, we compared our participants to the pretest sample on key demographic variables: age, gender, education and political ideology. The analyses indicated a marginal difference in the gender distribution, χ2 = 4.98 (2), p = 0.08; there were a greater proportion of women in the main study compared to the pretest. In addition, women rated the control unrelated conspiracy theories as more believable than the control related, t(159) = -2.77, p<0.01, d=0.21. When we re-ran the repeated measures ANOVA, with gender as a covariate, the main effect of conspiracy theory type was not present, F (1, 277) = 0.045, p =0.83, f = 0.01 and neither were the main effect of group, F (2, 277) = 0.225, p = 0.8, f = 0.04, nor the interaction effect F (2, 277) = 1.15, p = 0.318, f = 0.09. Thus, the significant effect of conspiracy theory type was most likely due to the higher proportion of women in the main study.

Study 6

Another plausible moderating factor determining the value of conspiracy theories for restoring threatened control, is the availability of alternative means of control. As noted above, while compensatory control processes do not require that sources of control be benevolent, people unsurprisingly favour sources of control that are not themselves threatening in other ways. Furthermore, there is a plethora of alternative control and

6 The pattern of results was unchanged when we analyzed the low control vs. neutral groups only. 118

meaning-making systems that are proven substitutes for personal control, such as God

(Laurin, Kay, & Moscovitch, 2008), government institutions (Shepherd et al., 2011), scientific theories (Rutjens, van Harreveld, van der Pligt, Kreemers & Noordevier, 2013), and even more abstract beliefs such as in meritocracy (Goode, Keefer, & Molina, 2014) and societal progress (Rutjens, van Harreveld & van der Pligt, 2010, 2013) . Even if these alternatives are not all equally effective at establishing a sense of control, conspiracy theories would seem to be an option of last resort, given the essentially malevolent worldview they imply, and the social stigma that can accompany their endorsement (Lantian, Muller, Nurra,

Klein, Berjot & Pantazi, 2018). Consequently, conspiracy theorizing may be a secondary strategy for restoring control, avoided when one has other means of affirming that the world is ordered (Rutjens, van der Pligt, & van Harreveld, 2010).

Thus, in Study 6, we tested the hypothesis that threats to control would increase belief in conspiracy theories to a greater degree (or only) when the effectiveness of a second salient source of control – the government (Kay & Sullivan, 2013) – was called into question.

Method

Participants. Six hundred and five MTurk volunteers participated (267 male, 337 female, 1 “other”). The majority (42.6 %) had an undergraduate degree, 24.5% had a high school degree, while 21.3% had a graduate degree, 9.3% an “other” degree and 2.3% had no degree. The mean age was 33.96 years (SD= 11.11 age range 18 – 78). None had participated in our previous studies.

Materials and Procedure. Participants were randomly assigned to a low control or a neutral group, as described in previous studies, and, independently, to a “competent government” or “incompetent government” condition. In the former, participants read a passage (about which they expected to answer questions) describing the U.S. government’s

119

response, generally considered competent and effective (Kluger & Sweetland Edwards,

2017), to Hurricane Irma which most severely struck Florida, Georgia and South Carolina in

2017; in the latter, they read a similar passage about the government’s response to Hurricane

Katrina, generally considered incompetent and ineffective (Committee on Homeland Security and Governmental Affairs, 2006), which struck Florida and Louisiana in August 2005. For example, Hurricane Irma’s response was described as “timely, well managed, planned and coordinated. Residents were ordered to a shelter of last resort with sufficient provision of food, water, security and sanitary conditions. No one was thirsty, exhausted or victim to violence.” Hurricane Katrina’s response was described as “delayed, mismanaged, unprepared and uncoordinated. Residents were ordered to a shelter of last resort without provision of adequate food, water, security or sanitary conditions. Several citizens died of thirst, exhaustion and violence, days after the storm had passed.” (The full text of both scenarios appear in the Appendix.)

Afterwards, consistent with the cover story, participants answered several multiple choice questions about the scenario (which were not analyzed). Afterwards, all participants completed the Conspiracy Mentality Scale, as described in Study 1, including an attention check, followed by demographic questions and self-assessment of data quality.

Results and Discussion

One hundred and four participants were excluded from the main analysis: 74 due to failures of attention and 30 due to failure to following instructions. This left a total of 501 participants. A sensitivity analysis (α = 0.05, 1 - β = 0.8) indicated that the study is able to detect a minimum effect size of f = 0.12.

A 2 (control priming: low vs. neutral) x 2 (government competence: competent versus incompetent) ANOVA on conspiracy theory ideation revealed no main effect of priming, F

(1, 497) = 0.161, p = 0.689, f = 0.02 or of competence, F (1, 497) = 0.243, p = 0.623, f =

120

0.02. Importantly, there was no interaction F (1, 497) =1.301, p = 0.255, f = 0.05, (see Table

2 for descriptive statistics). Thus, the study, again, provides no evidence that control priming influences conspiracy beliefs, regardless of whether an alternative source of control has been compromised.

Table 2. Descriptive statistics by recall task and scenario type.

Control priming Scenario Mean SD N

Incompetent 4.16 1.23 109 Low control Competent 3.98 1.39 125

Incompetent 4.08 1.25 133 Neutral Competent 4.15 1.07 134

Incompetent 4.11 1.24 242 Total Competent 4.06 1.23 259

Meta-analyses

Although Studies 1-6 varied somewhat in methodology, their consistent inclusion of low control and neutral conditions permitted a meta-analysis of the effects of control on conspiracy beliefs, which maximizes both power and accuracy of our effect size estimates.

Because in Study 2 the low control manipulation did not lower perceived control compared to the neutral group we excluded Study 2 from the analysis. For Studies 5 (low control compared to neutral) and 6 we looked at the effect at each level of the moderator. Because

121

the effect sizes in Studies 5 and 6 were not independent we used the robumeta (Fisher &

Tipton, 2015) package in R, which takes the dependency into account, to conduct the meta- analysis. A test of heterogeneity suggested consistency between the studies, I2 = 11. The cumulative effect size did not suggest effect of condition, d = -0.03, 95% CI [-0.20, 0.13], p

= 0.60 (see Figure 1).

Figure 1. Forest plot of analysed effect sizes from Studies 1and 3-6 (squares) and the cumulative effect size (diamond).

Bayesian analysis

To investigate support for the null hypothesis relative to the alternative hypothesis that control priming influences conspiracy theory beliefs, we used a Bayesian information criterion (BIC), using the calculator by Masson (2011), based on Wagenmakers (2007). The analysis requires transformation of the sum of squares generated by an ANOVA output and

122

provides graded level of evidence for the null and alternative hypothesis. According to

Raftery (1995), the strength of evidence for the posterior probability of a hypothesis given the data (pBIC(H1|D) within the range of 0.50-0.75 is “weak”, between 0.75-0.95 is “positive”, between 0.95-0.99 “strong” and above 0.99 “very strong”.

For those studies containing three groups we analysed the low and neutral (Study 1 and 5) or low and high (Study 2). For studies 5 and 6 we looked at each level of the moderator.

Table 3. shows the posterior probability of the null hypothesis given the data, pBIC(H0|D), and the probability of the alternative hypothesis given the data, pBIC(H1|D). There is positive evidence for the null hypothesis in five out of the six studies. In Study 1, there is only weak evidence for the null hypothesis, but the support for the alternative hypothesis is even lower (unsurprisingly, given that the effect is in the “wrong” direction).

Table 3. Overview of study design and posterior probabilities for H0 and H1.

Manipulation Manipulation DV pBIC(H0|D) pBIC(H1|D) check Study 1 Recall task ✓ CTI 0.60 0.40 Study 2 Modified recall task ✓ CTI 0.93 0.07 Study 3 Recall task CTI 0.90 0.10 Study 4 Recall task BCTI 0.93 0.07 Study 5 Recall task (Control related) Specific theories 0.93 0.07 Study 5 Recall task (Control non Specific theories 0.93 0.07 related) Study 6 Recall task (Irma Scenario) CTI 0.90 0.10 Study 6 Recall task (Katrina scenario) CTI 0.93 0.07 Note: CTI = The Conspiracy Theory Ideation Subscale from the Conspiracy Mentality Scale; BCTI = Beliefs in Conspiracy Theories Inventory

General Discussion

Six studies tested the hypothesis that lack of control motivates belief in conspiracies.

As argued in the Introduction, while the hypothesis is plausible, there is in fact little support

123

for it in the (small) literature on the subject. The current research offers no further evidence, and some positive evidence for the null hypothesis, at least when the most common experimental manipulation of control is used. Participants who were asked to recall instances in which they felt out of control were no more likely to endorse either general conspiracy beliefs, or specific conspiracy theories, relative to participants who recalled neutral memories, or nothing at all. Plausible moderating variables – the extent to which conspiracies provide evidence of control (Study 5), and the effectiveness of an alternative means of control (Study 6) – had no effect, and a meta-analysis of the studies determined a cumulative effect size not statistically different from zero. On the contrary, a Bayesian analysis indicated that the null hypothesis (i.e., that there is no causal relationship between lack of control and conspiracy theory beliefs) should be retained.

How to explain the inconsistency between previous positive results and our null findings? One explanation, of course, is that previous “positive” findings were simply Type 1 errors, which is not farfetched given social psychology’s historical inattention to power

(Clark‐Carter, 1997; Cohen, 1962). Another possibility, however, is that some previous studies measured constructs other than “pure” conspiracy ideation (e.g., pattern perception, paranoid beliefs, corruption beliefs), which are susceptible to control threats. Indeed, studies that examine more typical conspiracy theories and have larger samples tend to report null results. For example, Hart and Graether (2018), found no effects of control on the Generic

Conspiracist Beliefs Scale (another psychometrically validated measure of conspiracy beliefs, with items such as, “Experiments involving new drugs or technologies are routinely carried out on the public without their knowledge or consent”). Similarly, Nyhan and Zeitzoff (2018) found no effect of control priming on Middle Eastern and North African participants’ conspiracy beliefs, using a list of conspiracy theories consisting of items such as “Jewish

124

leaders are secretly conspiring to achieve world domination.” Thus, it seems that overall there is little empirical evidence causally linking lack of control and conspiracy beliefs.

Although there was no evidence in the current studies for a causal effect of control on conspiracy beliefs, Studies 1 and 2 provided evidence for a correlational relationship, consistent with previous research (e.g., Abalakina-Paap, Stephan, Craig, & Gregory, 1998;

Imhoff, 2015; Imhoff & Bruder, 2014; van Elk & Lodder, 2018; van Prooijen & Acker,

2015), suggesting that chronic, rather than acute feelings of control are related to conspiracy theory beliefs. While longitudinal studies may help shed light on the effects of chronic lack of control on conspiracy beliefs, it is also possible that the link is spurious. There are any number of co-variates of both lack of control and conspiracy theory belief, including stress

(Swami, Furhnam, Smyth, Weis, Lay & Clow, 2016; Seligman, 1975), self-esteem

(Cichocka, Marchlewska, & de Zavala, 2016; Ledrich, & Gana, 2013), anxiety (Lachman, &

Agrigoroaei, 2012; Grzesiak-Feldman, 2013 ) or uncertainty (van Prooijen & Jostmann,

2013) that could account for their association. Also, the correlation may be due to conspiracy beliefs’ shared variance with paranoia (see for example Imhoff & Lamberty, 2018).

Alternatively, the causality may run in the opposite direction: conspiracy theories themselves may reduce feelings of control. Jolley and Douglas (2014), for example, found that participants exposed to conspiracy theories about vaccines reported greater feelings of powerlessness compared to participants exposed to anti-conspiracy material, suggesting that conspiracy beliefs threaten rather than satiate one’s sense of control. In any case, it is clear that presence of correlational evidence in absence of experimental evidence is not a strong foundation for claiming that lack of control is precursor to conspiracy beliefs.

Even if these null results are confirmed in further research, however, it is important to acknowledge their limitations. Priming is only one method of threatening (or bolstering) control, and while we found no evidence for changes in conspiracy beliefs, stronger or

125

qualitatively distinct manipulations could conceivably produce effects that we were unable to detect here. Furthermore, we could not assess the effects of chronic lack of control: It may be that prolonged periods of control deprivation do indeed produce reliance on other, even malevolent sources of control, including conspiracy theories. Another unexplored area is whether lack of control in a specific domain, will lead to belief in conspiracy theory in that domain. For example, people diagnosed with HIV, might be particularly inclined to believe in conspiracy theories involving that virus, rather than conspiracy theories in general.

To conclude, no evidence in our studies supports the casual link from lack of control to conspiracy theory beliefs. Future studies could use alternative manipulations, measures, and designs (i.e. longitudinal) to explore the causality, if any, that accounts for the negative correlation.

126

References

Abalakina-Paap, M., Stephan, W. G., Craig, T., & Gregory, W. L. (1999). Beliefs in

conspiracies. Political Psychology, 20, 637-647. https://doi.org/10.1111/0162-

895X.00160

Brotherton, R. (2015). Suspicious minds: Why we believe conspiracy theories. London,

England: Bloomsbury.

Brotherton, R., & French, C. C. (2014). Belief in Conspiracy Theories and Susceptibility to the

Conjunction Fallacy. Applied Cognitive Psychology, 28, 238-248.

https://doi.org/10.1002/acp.2995

Brotherton R., & French, C. (2015). Intention Seekers: Conspiracist Ideation and Biased

Attributions of Intentionality. PLoS ONE 10(5): e0124125.

doi:10.1371/journal.pone.0124125

Chen, C. Y., Lee, L., & Yap, A. J. (2017). Control deprivation motivates acquisition of utilitarian

products. Journal of Consumer Research, 43, 1031-1047.

https://doi.org/10.1093/jcr/ucw068

Cichocka, A., Marchlewska, M., & de Zavala, A. G. (2016). Does Self-Love or Self-Hate Predict

Conspiracy Beliefs? Narcissism, Self-Esteem, and the Endorsement of Conspiracy

Theories. Social Psychological and Personality Science, 7, 157-166.

https://doi.org/10.1177/1948550615616170

Cichocka, A., Marchlewska, M., Golec de Zavala, A., & Olechowski, M. (2016). ‘They will not

control us’: Ingroup positivity and belief in intergroup conspiracies. British Journal of

Psychology, 107, 556–576.

Clark‐Carter, D. (1997). The account taken of statistical power in research published in

the British Journal of Psychology. British Journal of Psychology, 88, 71-83.

doi:10.1111/j.2044-8295.1997.tb02621.x

127

Clarke, S. (2002). Conspiracy Theories and Conspiracy Theorizing. Philosophy of the Social

Sciences, 32(2), 131–150. https://doi.org/10.1177/004931032002001

Cohen, J. (1962). The statistical power of abnormal-social psychological research: A review. The

Journal of Abnormal and Social Psychology, 65(3), 145-153.

Committee on Homeland Security and Governmental Affairs. (2006). Hurricane Katrina: a

nation still unprepared : special report of the Committee on Homeland Security and

Governmental Affairs, United States Senate, together with additional views. Washington

:U.S. G.P.O. Available at: https://www.govinfo.gov/content/pkg/CRPT-

109srpt322/pdf/CRPT-109srpt322.pdf

Cutright, K. M. (2012). The Beauty of Boundaries: When and Why We Seek Structure in

Consumption. Journal of Consumer Research, 38, 775-790.

https://doi.org/10.1086/661563

Cutright, K. M., Bettman, J. R., & Fitzsimons, G. J. (2013). Putting Brands in Their Place: How a

Lack of Control Keeps Brands Contained. Journal of Marketing Research, 50, 365-378.

https://doi.org/10.1509/jmr.10.0202

Douglas, K.M., & Sutton, R.M. (2018). Why conspiracy theories matter: A social psychological

analysis. European Review of Social Psychology, 29, 256 298.

Douglas, K. M., Sutton, R. M., & Cichocka, A. (2017). The psychology of conspiracy theories.

Current Directions in Psychological Science, 26, 538-542.

Douglas, K., Sutton, R., Jolley, D., & Wood, M. (2015). The social, political, environmental, and

health-related consequences of conspiracy theories: problems and potential solutions. In

M. Bilewicz, A. Cichocka, & W. Soral (Eds.), The psychology of conspiracy (pp.183-

196). East Sussex: Routledge.

128

Douglas, K.M., Sutton, R.M., Callan, M.J., Dawtry, R.J., & Harvey, A.J. (2016). Someone is

pulling the strings: hypersensitive agency detection and belief in conspiracy theories.

Thinking & Reasoning, 22, 57-77. https://doi.org/10.1080/13546783.2015.1051586

Fisher, Z., & Tipton, E. (2015). Robumeta: An R-package for robust variance estimation in meta-

analysis. Retrieved from: https://arxiv.org/abs/1503.02220)

Goode, C., Keefer, L. A., & Molina, L. E. (2014). A Compensatory Control Account of

Meritocracy. Journal of Social and Political Psychology, 2, 313-334.

https://doi.org/10.5964/jspp.v2i1.372

Goode, C., Keefer, L. A., Branscombe, N. R., & Molina, L. E. (2017). Group identity as a source

of threat and means of compensation: Establishing personal control through group

identification and ideology. European Journal of Social Psychology, 47(3), 259-272.

doi:10.1002/ejsp.2259

Green, R., & Douglas, K. M. (2018). Anxious attachment and belief in conspiracy theories.

Personality and Individual Differences, 125, 30-37.

ttps://doi.org/10.1016/j.paid.2017.12.023

Grzesiak-Feldman, M. (2013). The effect of high-anxiety situations on conspiracy thinking:

Research and reviews research and reviews. Current Psychology, 32(1), 100-118.

doi:http://dx.doi.org.ezproxy.otago.ac.nz/10.1007/s12144-013-9165-6

Hart, J. & Graether, M. (2018). Something’s Going on Here. Psychological Predictors of Belief

in Conspiracy Theories. Journal of Individual Differences, 39 (4), 229-237.

Hauser, D. J., Ellsworth, P. C., Gonzalez, R. (2018). Are manipulation checks

necessary? Frontiers in Psychology, 9: 998. doi:10.3389/fpsyg.2018.00998

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.

129

Imhoff, R. (2015). Beyond (right-wing) authoritarianism: Conspiracy mentality as an incremental

predictor of prejudice. In M. Bilewicz, A. Cichocka & W. Soral (Eds.) The Psychology of

Conspiracy Theories (pp.122-141). East Sussex: Routledge.

Imhoff, R., & Bruder, M. (2014). Speaking (Un-)truth to power: Conspiracy mentality as a

generalised political attitude. European Journal of Personality, 28, 25-43.

https://doi.org/10.1002/per.1930

Imhoff, R., & Lamberty, P. (2018). How paranoid are conspiracy believers? Toward a more fine‐

grained understanding of the connect and disconnect between paranoia and belief in

conspiracy theories. European Journal of Social Psychology, 48(7), 909-926.

https://doi.org/10.1002/ejsp.2494

Jolley, D., & Douglas, K.M. (2014). The Effects of Anti-Vaccine Conspiracy Theories on

Vaccination Intentions. PLoS ONE 9(2): e89177.

https://doi.org/10.1371/journal.pone.0089177

Kay, A. C., & Eibach, R. P. (2013). Compensatory control and its implications for ideological

extremism. Journal of Social Issues, 69(3), 564-585. doi:10.1111/josi.12029

Kay, A. C., Gaucher, D., McGregor, I., & Nash, K. (2010). Religious Belief as Compensatory

Control. Personality and Social Psychology Review, 14(1), 37–48.

https://doi.org/10.1177/1088868309353750

Kay, A. C., Gaucher, D., Napier, J. L., Callan, M. J., & Laurin, K. (2008). God and the

Government : Testing a Compensatory Control Mechanism for the Support of External

Systems. Journal of Personality and Social Psychology, 95, 18–35.

https://doi.org/10.1037/0022-3514.95.1.18

Kay, A.C. & Sullivan, D. (2013). Cultural unity and diversity in compensatory control processes.

In M.J. Gelfand, C. Chiu, & Y. Hong (Eds.), Advances in culture and psychology (Vol. 3,

pp. 181-226). New York: Oxford University Press.

130

Kay, A. C., Sullivan, D., & Landau, M. J. (2015). Psychological importance of beliefs in control

and order: Historical and contemporary perspectives in social and personality psychology.

In M. Mikulincer, P. R. Shaver, E. Borgida, & J. A. Bargh (Eds.), APA handbooks in

psychology. APA handbook of personality and social psychology, Vol. 1. Attitudes and

social cognition (pp. 309-337). Washington, DC, US: American Psychological

Association. http://dx.doi.org/10.1037/14341-010

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. https://doi.org/10.1111/j.1467-8721.2009.01649.x

Keeley, B. (1999). Of Conspiracy Theories. Journal of Philosophy 96, 109–26.

https://doi.org/10.2307/2564659

Kluger, J. & Sweetland Edwards, H. (2017, September 27). The Angels of Irma. Time. Retrieved

from: https://time.com/4941014/the-angels-of-irma/

Kossowska, M., & Bukowski, M. (2015). Motivated roots of conspiracies: The role of certainty

and control motives in conspiracy thinking. In M. Bilewicz, A. Cichocka & W. Soral

(Eds.) The Psychology of Conspiracy Theories (pp. 145-161). East Sussex: Routledge.

Lachman, M. E., & Agrigoroaei, S. (2012). Low perceived control as a risk factor for episodic

memory: The mediational role of anxiety and task interference.Memory &

Cognition, 40(2), 287-296. doi:http://dx.doi.org.ezproxy.otago.ac.nz/10.3758/s13421-

011-0140-x

Landau M. J., Kay A. C., Whitson J. A. (2015). Compensatory control and the appeal of a

structured world. Psychological Bulletin, 141, 694–722.

Lantian, A., Muller, D., Nurra, C., Klein, O., Berjot, S., & Pantazi, M. (2018). Stigmatized

beliefs: Conspiracy theories, anticipated negative evaluation of the self, and fear of social

131

exclusion. European Journal of Social Psychology, 48 (7), 939-954.

https://doi.org/10.1002/ejsp.2498

Laurin, K., Kay, A. C., & Moscovitch, D. A. (2008). On the belief in God : Towards an

understanding of the emotional substrates of compensatory control. Journal of

Experimental Social Psychology, 44, 1559–1562.

https://doi.org/10.1016/j.jesp.2008.07.007

Ledrich, J.& Gana, K. (2013). Relationship between attributional style, perceived control, self‐

esteem, and depressive mood in a nonclinical sample: A structural equation‐modelling

approach. Psychology and Psychotherapy: Theory, Research and Practice, 86, 413-430.

doi:10.1111/j.2044-8341.2012.02067.x

Ma, A., & Kay, A. C. (2017). Compensatory control and ambiguity intolerance. Organizational

Behavior and Human Decision Processes, 140, 46-61. doi:10.1016/j.obhdp.2017.04.001

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, 109–117. https://doi.org/10.1002/ejsp.2308

Mashuri, A. & Zaduqisti, E. (2015). The effect of intergroup threat and social identity salience on

the belief in conspiracy theories over terrorism in Indonesia: Collective angst as mediator.

International Journal of Psychological Research, 8, 24-35.

Mashuri, A., & Zaduqisti, E. (2014). The role of social identification, intergroup threat, and out-

group derogation in explaining belief in conspiracy theory about terrorism in

Indonesia. International Journal of Research Studies in Psychology, 3(1), 35-50.

DOI: 10.5861/ijrsp.2013.446

Masson, M. E. J. (2011). A tutorial on a practical Bayesian alternative to null-hypothesis

significance testing. Behavior Research Methods, 43, 679–690.

https://doi.org/10.3758/s13428-010-0049-5

132

Nyhan, B., & Zeitzoff, T. (2018). Conspiracy and Misperception Belief in the Middle East and

North Africa. The Journal of Politics, 80(4), 1400-1404.

Oaklander, M (2015, August). Here’s why people believe in conspiracy theories. Time. Retrieved

from: http://time.com/3997033/conspiracy-theories/

Pearlin, L. I., Menaghan, E. G., Lieberman, M. A., & Mullan, J. T. (1981). The Stress Process.

Journal of Health and Social Behavior, 22, 337-356. https://doi.org/10.2307/2136676

Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of Health and Social

Behavior, 19(1), 2-21. http://dx.doi.org/10.2307/2136319

Pichaske, K. (Producer) & Alpert, D. (Producer/Director) (2018). Pulling the Thread. United

States: Kindling Group.

Raftery, A. E. (1995). Bayesian model selection in social research. Sociological methodology,

25, 111-163. https://doi.org/10.2307/271063

Richard, F. D., Bond Jr., C. F., & Stokes-Zoota, J. J. (2003). One hundred years of social

psychology quantitatively described. Review of General Psychology, 7(4), 331-363.

doi:10.1037/1089-2680.7.4.331

Rutjens, B.T., & Kay, A. (2016). Compensatory control theory and the psychological importance

of perceiving order. In M. Bukowski, I. Fritsche, A. Guinote, and M. Kofta (Eds.) Coping

with a lack of control in a social world (pp. 83-96). London: Routledge

Rutjens, B. T., van Harreveld, F. & van der Pligt, J. (2010). Yes We Can: Belief in Progress as

Compensatory Control. Social Psychological and Personality Science, 1(3), 246–

252. https://doi.org/10.1177/1948550610361782

Rutjens, B. T., van Harreveld,, van der Pligt, J., Kreemers, L.M., & Noordewier, M.K. (2013).

Steps, Stages, and Structure : Finding Compensatory Order in Scientific Theories.

Journal of Experimental Social Psychology, 142, 313–318.

https://doi.org/10.1037/a0028716

133

Rutjens, B. T., van Harreveld, F., & van der Pligt, J. (2013). Step by Step: Finding Compensatory

Order in Science. Current Directions in Psychological Science, 22(3), 250–

255. https://doi.org/10.1177/0963721412469810

Rutjens, B. T., van der Pligt, J., & van Harreveld, F. (2010). Deus or Darwin: Randomness and

belief in theories about the origin of life. Journal of Experimental Social Psychology, 46,

1078-1080. https://doi.org/10.1016/j.jesp.2010.07.009

Shermer, M. (2014, December). Conspiracy Central. Scientific American, 311 (6), 94.

Seligman, M. (1975). Helplessness: on depression, development, and death. San Francisco : W.

H. Freeman.

Shepherd, S., Kay, A. C., Landau, M. J., & Keefer, L. A. (2011). Evidence for the specificity of

control motivations in worldview defense : Distinguishing compensatory control from

uncertainty management and terror management processes. Journal of Experimental

Social Psychology, 47, 949–958. https://doi.org/10.1016/j.jesp.2011.03.026

Stojanov, A. & Halberstadt, J. (2019). The conspiracy mentality scale: Distinguishing between

irrational and rational suspicion. Social Psychology. Advanced Online Article.

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., Furnham, A., Smyth, N., Weis, L., Lay, A., & Clow, A. (2016). Putting the stress on

conspiracy theories: Examining associations between psychological stress, anxiety, and

134

belief in conspiracy theories. Personality and Individual Differences, 99, 72-76.

https://doi.org/10.1016/j.paid.2016.04.084 van Elk, M. & Lodder, P. (2018). Experimental Manipulations of Personal Control do Not

Increase Illusory Pattern Perception. Collabra, 4, 1–22.

Van Prooijen, J.-W. (in press). An existential thread model of conspiracy theories. European

Psychologist. van Prooijen, J.-W. (2018). The psychology of conspiracy theories. London: Routledge. van Prooijen, J-W., & Acker, M. (2015). The influence of control on belief in conspiracy

theories: Conceptual and applied extensions. Applied Cognitive Psychology, 29, 753-761. van Prooijen, J.‐W., Douglas, K. M., & De Inocencio, C. (2018). Connecting the dots: Illusory

pattern perception predicts belief in conspiracies and the supernatural. European Journal

of Social Psychology, 48, 320–335. 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(1), 109-

115. http://dx.doi.org/10.1002/ejsp.1922

Wagenmakers, E. J. (2007). A practical solution to the pervasive problems of p values.

Psychonomic Bulletin and Review, 14, 779-804. https://doi.org/10.3758/BF03194105

Whitson, J. A., & Galinsky, A. D. (2008). Lacking control increases illusory pattern perception.

Science, 322, 115-117. https://doi.org/10.1126/science.1159845

Wood, M.J. & Douglas, K.M. (2019). Conspiracy Theory Psychology: Individual Differences,

Worldviews and States of Mind. In J.E. Uscinski (Ed.) Conspiracy Theories and the

People Who Believe Them (pp. 245-256). New York, NY: Oxford University Press.

Yong, E. (2008, December). Lacking control drives false conclusions, conspiracy theories and

superstitions. National Geographic. Retrieved from:

135

https://www.nationalgeographic.com/science/phenomena/2008/12/27/lacking-control- drives-false-conclusions-conspiracy-theories-and-superstitions/

136

Appendix

These scenarios were adapted from information available at Wikipedia and

https://www.cato.org/blog/hurricane-katrina-remembering-federal-failures

Hurricane Katrina scenario

The government has come under intense criticism in recent years for its ability to carry out its key functions. Nowhere was this better exemplified than by its response to

Hurricane Katrina.

The government response to hurricane Katrina was delayed, mismanaged, unprepared and uncoordinated.

Residents were ordered to a shelter of last resort without provision of adequate food, water, security or sanitary conditions. Several citizens died of thirst, exhaustion and violence, days after the storm had passed.

Indecision plagued government leaders in the deployment of supplies, in medical personnel decisions, and in other areas. In places that desperately needed help, it took days to deliver medical supplies. Even the grisly task of body recovery after Katrina was slow and confused. Bodies went uncollected for days.

Further, federal agencies were unfamiliar with their roles and responsibilities. There was general confusion over mission assignments, deployments, and command structure.

Agencies could not communicate with each other due to equipment failures and a lack of system interoperability. The Federal Emergency Management Agency (FEMA) was not able to coordinate its efforts with other federal agencies.

To conclude, the response to hurricane Katrina nicely illustrates government inefficiency at all levels, and serves as a reminder of its chaotic functioning.

137

Hurricane Irma Scenario

The government has come under intense criticism in recent years for its ability to carry out its key functions, but in fact the government has shown remarkable competence when it is needed. Nowhere was this better exemplified than by its response to Hurricane

Irma.

The government response to hurricane Irma was timely, well managed, planned and coordinated.

Residents were ordered to a shelter of last resort with sufficient provision of food, water, security and sanitary conditions. No one was thirsty, exhausted or victim to violence.

Government leaders were decisive in deployment of supplies, in medical personnel decisions, and in other areas. In places that desperately needed help, medical supplies were quickly delivered. The body recovery after Irma was fast and organized.

Further, federal agencies were familiar with their roles and responsibilities. Mission assignments, deployments, and command structure were clear.

System interoperability and effective equipment allowed agencies to communicate with each other. The Federal Emergency Management Agency (FEMA) effectively coordinated its efforts with other federal agencies.

To conclude, the government response to hurricane Irma nicely illustrates government efficiency at all levels, and serves as a reminded of its orderly functioning.

138

Online Supplemental Material

Vera Vera Protot Protot

city city ypical ypical

M SD ity M ity SD

A family friendly entertainment site that serves pizza is a cover up for Hillary Clinton's child sex ring 1.94 2.03 4.89 3.19

A powerful secret elite with a sinister agenda is ruling the world 3.71 2.83 6.24 2.64

Acts that victimize members of a minority race are specifically planned and targeted 4.14 2.7 5.02 2.73

Aliens are behind the world problems and disasters 1.49 1.11 4.89 3.04

An alien race of lizard shapeshifters actually controls the top government positions 1.81 1.72 5 3.26

An alien spacecraft crashed in Roswell, New Mexico and the government covered it up 3.47 2.74 6.47 2.85

Animal advocates fabricate the factory farming industry's mistreatment of animals to try to sway people into not getting the nutrients they need from animal products 2.29 1.96 4.88 2.76

Bigfoot, a human animal hybrid, is a scientific experiment gone wrong 1.73 1.37 4.98 2.94

Billionaires have the presidents of the US on their payroll so that they do exactly what they want them to 4.09 2.84 6.48 2.5

Companies purposely products that do more harm than good and create a vicious cycle for consumers to buy more and more 4.36 2.84 5.81 2.33

Corporations and superrich oil dynasties are buying up water rights, controlling nations and populations 4.94 2.71 6.02 2.63

139

Cures for diseases such as Cancer and AIDS exist, but are not made public so that the population can be controlled 3.75 3.1 6.28 2.81

Dangerous chemical trails are left in the air by jets in the sky to purposely harm people and keep them sick and docile 2.59 2.44 5.73 3.03

Democrats are spreading rumours that Trump has interests and ties to Russia as an attempt to stop Trump from undoing Obama's policies 3.33 2.84 5.55 3.13

Dinosaurs never existed Carbon dating doesn't exist God did this to test our faith 1.73 1.91 4.15 3.26

Donald Trump only ran for president because it would boost his media presence and allow him to gain more money out of his TV show, but he never thought he would actually win 5.43 2.91 4.75 2.91

Edward Snowden is actually a part of complicated NSA plot, and the US spy programme was purposely exposed so that citizens would fear the NSA 2.42 2.02 5.51 3.05

Evidence supporting the safety and benefit of genetically modified food crops is fabricated, while evidence suggesting the dangers of these crops is suppressed 4.09 2.94 5.94 2.46

George W Bush is a robot 1.4 1.29 3.55 3.14

Global warming is made up and is a way to bring money to a certain group of people 2.07 2.03 5.46 3.11

Governments use "vaccinations" as disguises to put microchips into citizens so they can track their movement 1.92 1.74 5.66 3.14

Hillary Clinton is a spy for other governments and stole secret information via email 1.97 2 5.18 2.99

140

HIV was manmade in order to control the population 2.7 2.53 5.89 3.09

In the 1969 IsraeliArab war, the Israelis attacked USS Liberty on purpose, and not, as they claimed, by accidentally mistaking it for an Egyptian ship 2.84 2.29 6.04 2.64

Iran is the actual money and muscle behind Al Qaeda 3.57 2.33 5.55 2.61

Jews poisoned wells and consumed the blood of Christians 1.56 1.23 3.74 3.01

Jews secretly control the banking system and conspire for nefarious purposes 2.21 2.24 5.44 3.15

Kale is pushed to be marketed over other more nutritious vegetables, resulting in the hype 3.44 2.61 4.4 2.91

Kim Jung Un doesn't bleed, and can't actually die 1.35 1.22 3.31 3.09

Lyme disease is a cross between aids and syphilis that escaped the lab via an infected tick 1.92 1.83 4.69 2.98

Marilyn Monroe was murdered because she knew too much due to her affair with President Kennedy 4.1 2.86 6.62 2.57

Michael Jackson was the same person as La Toya Jackson 1.52 1.45 3.78 3.03

Michelle Obama is a transgender woman who started out as a man 1.73 1.73 4.18 3.16

NASA faked the 1969 moon landing 2.17 2.17 6.13 3.04

New age belief systems are trying to debunk God's existence 3.84 2.76 4.18 2.99

O.J. Simpson was framed for the murders of his ex-wife and her friend. 2.04 1.94 4.66 2.96

Obama wanted to attack and start a Texas civil war during the summer of 2015 1.64 1.39 4.21 3.14

141

Osama Bin Laden died just before 9/11 but was used as a scapegoat 2.27 2.11 5.28 3.12

Paid operatives cast votes for the dead people. 3.55 2.73 5.55 2.75

Paul McCartney died and was replaced by a look alike 1.65 1.52 4.31 3.17

People that have knowledge of illegal government activities are being killed and disposed of. 4.57 2.82 6.13 2.53

Planned Parenthood encourages abortions so they can make money off of selling body parts of dead babies. 2.74 2.69 4.95 3.18

Planned Parenthood is seeking to destroy the black population by injecting them with lethal drugs 1.67 1.54 5 3.28

Pop star Katy Perry is really JobBenét Ramsey, the sixyearold beauty pageant queen who was murdered in the 90s 1.36 1.07 4.12 3.19

President Obama is secretly a Muslim who wasn't born in Hawaii 2.14 2.33 5.39 3.14

President Obama was wiretapping President Trump's phone lines in the days prior to the election 2.82 2.53 5.6 2.86

President Roosevelt knew that Pearl Harbor was going to be bombed by the Japanese, but did not stop it because he knew it would force the United States to enter World War II 3.42 2.64 5.54 2.81

Princess Diana was not killed in an ordinary car crash; it was an assassination 3.66 2.75 6.65 2.59

Public schools teach specifically to indoctrinate and brainwash students with the desired culture 4.08 2.92 5.25 2.87

Queen Elizabeth died at age 12 and was replaced with a boy That is why the queen never married nor appeared in public without makeup 1.67 1.34 4.86 3.31

142

Russia sent numerous humans into space before Yuri Gagarin, but they died there 2.61 1.94 5.18 2.63

Russia started the hippie movement in the US because it wanted to undermine and weaken America's strong traditional values 2.23 2.23 5.18 3.08

Russians believe that only Americans can assassinate the

American way of life 3.27 2.18 3.68 2.83

Shakespeare didn't write his works; they were written by another author 2.72 2.11 4.76 2.96

Some foods are too big as a result of overgrowing them by the use of steroids 4.07 2.96 4.24 2.48

Some gamblers and some members of the Indianapolis Colts were paid to lose the 1969 Superbowl 2.73 2.04 5.68 2.72

Soy is being used to feminize men 1.97 1.87 4.74 2.99

Stevie Wonder has been faking his disability 1.58 1.48 4.28 3.09

Supreme Court Justice Scalia was murdered so that Obama could nominate a liberal leaning replacement 2.07 1.93 5.71 2.88

The 2004 Indian Ocean tsunami was intentionally caused by a nuclear weapon detonated in a strategic position under the ocean 1.93 1.71 5.21 3.13

The 2017 elections were rigged 4.49 3.1 6.31 2.63

The cameras on phones, TVs, computers and laptops are accessed by third party government agencies that monitor citizens and watch for illicit activity 4.2 2.83 5.98 2.56

The Clintons have had people killed to keep them quiet or to tie up loose ends for some of their activities that are extremely illegal 3.3 2.85 6.14 2.89

143

The criminal charges against Hilary Clinton were all due to a conspiracy from the right to make her look bad 3.92 2.88 5.25 2.96

The Democrats and the Republicans really have the same secret agenda and only appear as two choices so the American people think they have a choice 3.67 2.84 5.61 2.8

The development of alternative fuel solutions has been suppressed in order to avoid huge monetary losses from decreased use of fossil fuels 5.2 2.85 6.25 2.42

The events in the Bible didn’t really happen but were made up 4.29 3.04 3.54 2.72

The fat/oil/sugar manufacturers push nonnutritious food and poison an ever bloated, waddling public to pharmaceuticals 4.17 2.79 5.09 2.65

The government and NASA trick people into thinking the Earth is round when it's actually flat 1.44 1.4 4.59 3.28

The government deliberately flooded Louisiana to save the homes of the wealthy 2.31 2.22 5 3.27

The government planted cocaine in black neighbourhoods to keep black society addicted and helpless 3.27 2.71 5.11 3.13

The government possesses a machine that can generate natural disasters 2.19 2.09 4.86 3.06

The government puts fluorine into our water supply in order to make us subordinate 2.4 2.26 5.74 2.9

The government uses media and the entertainment industry to control minds 4.35 3.03 6.24 2.54

The Holocaust is made up just so Jewish people get better treatment from society 1.56 1.49 4.31 3.27

144

The identification chips implanted into household pets are being used to spy on citizens 2.26 1.94 5.58 2.84

The Jews in Israel are fake Jews The true Jews are the Black

Americans 1.48 1.49 3.36 2.92

The Mafia Killed Jimmy Hoffa to cover up their influence over the Unions 4.45 2.81 5.94 2.52

The media falsely reports the demographics of the population to push an agenda for homosexuals 2.68 2.52 4.85 2.96

The media is covering up that the Father of Ted Cruz is linked to

Kennedy assassination 2.49 2.23 5.29 3.07

The media is steering the narrative against the current

Washington administration by deliberately faking news stories 3.23 2.87 5.89 2.74

The moon is actually a large spaceship projecting a "matrix "to appear like a rocky surface 1.51 1.36 4.54 3.25

The Muslims are taking over the world 2.7 2.28 4.34 3.03

The push for all cable to be broadcast digitally and for analog to be done away with was a conspiracy to get everyone properly hooked up with the necessary equipment for the government to spy on citizens 3.02 2.6 5.86 2.98

The pyramids in Egypt were built by aliens 2.08 2.06 5.18 3.23

The rapper Tupac Shakur is still alive and wasn't shot and killed in the 90's 2.02 2 5.18 3.15

The Sandy Hook shootings did not really happen; it was a hoax designed by the Government to gather support for stricter gun control laws 2.22 2.47 5.66 3.3

145

The taxes are not going completely to the people but for politicians’ own personal gain 5.11 2.96 5.47 2.65

The US didn’t attack Saddam Hussein because they suspected weapons of mass destruction but to eliminate a rival power to

Israel 3.32 2.7 5.71 2.63

The US economy is really controlled by the Russians and

Chinese 2.85 2.21 5.71 2.72

The US Government hid the existence of soldiers left behind during the Vietnam war They were then taken to labour camps in

Russia 3 2.33 5.73 2.81

The US Government was complicit in the 9/11 terrorist attacks 3.36 2.6 6.42 2.69

The US wants Assad gone in order to put a gas pipeline through

Syria 3.4 2.63 5.39 2.76

The USSR only appeared to collapse to cause a false sense of safety in the West 2.32 1.97 4.92 2.83

The weather is controlled by multiple governments to help various industries and hurt others 1.86 1.8 4.99 3.12

The Y2K bug was just a scheme by computer programmers and businesses to make money 3.68 2.87 5.66 2.72

There are aliens walking among us right now 2.81 2.49 5.07 3.03

There are magical forces that make ships disappear in the

Bermuda triangle 2.41 2.1 4.6 3.02

There are people that are time travellers who can often be seen in old photos 2.24 2.15 4.46 3.16

146

There are secret tunnels connecting Denver International Airport to NORAD 3.33 2.37 5.21 2.9

There is a pit on Oak island, off the coast of Canada, that people have been digging for centuries but the hole is boobytrapped with tunnels that let in water from all over the island the further down you dig 3.11 2.42 4.27 2.95

There is life on Mars, but the Government wants us to believe there isn’t 2.5 2.1 5.31 3

There is secret alien base on antarctica 2.21 2.04 4.89 3.24

There was a conspiracy behind the murder of JFK 5.08 2.78 6.84 2.44

Through several shell companies, the Clinton Foundation acquired and supplied Iran with Uranium in exchange for funds 2.94 2.51 5.88 2.71

Timothy McVeigh didn't blow up the Oklahoma buildings. He was a government patsy and instead of receiving the death penalty lives in some island far away 2.01 1.74 5.38 3.09

Trump is conspiring with Russia to start a war 3.61 2.77 5.47 2.88

TWA Flight 800 was actually shot down mistakenly by a missile from a US warship and did not crash – as official reports claim – as a result of an explosion of flammable fuel/air vapors in the fuel tank 3.09 2.32 6.29 2.51

Vaccines are a giant conspiracy by "Big Pharma" to stifle children and make them easily pliable so that the government can mold them into soulless automatons 2.74 2.65 5.75 3.08

Walt Disney hid immoral subliminal messages in his films 2.92 2.55 5.16 3.03

147

Wars are started and maintained by those who benefit from them

(ie arms manufacturers, oil companies etc) 4.94 2.88 5.91 2.74

World War II never occurred, it was simply a fiction planned by world leaders to keep their citizens in line 1.3 0.92 4.22 3.38

148

Chapter 4: Does lack of control lead to conspiracy beliefs? A

meta-analysis

149

Does lack of control lead to conspiracy beliefs? A meta-analysis

Ana Stojanov, 1 Jamin Halberstadt 1

1 Department of Psychology, University of Otago, Dunedin, New Zealand

Abstract

Perceived lack of control is widely believed to motivate, at least partly, belief in conspiracy theories. We question the theoretical foundations of this belief and meta-analyze existing published and unpublished studies to assess the overall effect of lack of control on conspiracy beliefs. The overall effect was small and not statistically significant (d = - 0.05), and was not moderated by comparison group (baseline vs control affirmation), type of manipulation used to threaten control, inclusion of a manipulation check, or sample type.

However, the predicted effect of control was more likely to be observed when beliefs were measured in terms of specific conspiracy theories, rather than as general or abstract claims.

Overall, the present studies to date offer limited support for the hypothesis that conspiracy beliefs arise as a compensatory control.

Keywords: conspiracy theories, compensatory control, lack of control, meta-analysis

150

Conspiracy theories, such as the belief that a secretive organization called the New

World Order is ruling the world from the shadows, or that the sky is sprayed with harmful chemicals, are becoming increasingly prominent (Collins, 2018). The reasons are multiple: the Internet has vastly increased the ease with which information can be spread, without a corresponding increase in the vetting of information for accuracy (Sharma, Yadav, Yadav &

Ferdinand, 2017); the structure and business models of many social media companies limit exposure to disconfirming information; popular books and movies have shifted fringe theories to the mainstream and increased their social acceptability (Kelley-Romano, 2008); trust in authority and expertise has eroded (Stevens, 2019; Tsipursky, 2018).

Conspiracy beliefs’ migration from the periphery to the mainstream has, however, been accompanied by a growing awareness among researchers of their potentially harmful effects, both for individuals and societies (Jolley, Douglas, Leite, & Schrader, 2019; Jolley,

Meleady & Douglas, 2019; Douglas & Sutton, 2015; 2018; Bilewicz, Winiewski, Kofta &

Wójcik, 2013), and a growing interest in understanding them. Some researchers have explored the motivational roots of conspiracy theory beliefs (Douglas, Sutton, & Cichocka,

2017), proposing links, for example, with the need to see the world as structured and ordered

(Landau, Kay & Whitson, 2015), an idea crystalized in compensatory control theory (Kay,

Gaucher, Napier, Callan & Laurin, 2008). According to the theory, when personal control, people’s primary means to perceive the world as ordered and non-random, is frustrated, they will resort to compensatory means of re-establishing order, such as believing in a controlling

God (Laurin, Kay, & Moscovitch, 2008), a highly competent government (Shepherd, Kay,

Landau, & Keefer, 2011), or even more abstract beliefs such as in meritocracy (Goode,

Keefer, & Molina, 2014). From this perspective, despite their often malevolent assumptions, conspiracy theories are an effective means of making sense of events that threaten order

(Whitson & Galinsky, 2008). The idea has gained traction among researchers and even in

151

popular culture, featuring recently in documentary series (Pichaske & Alpert, 2018), books

(Brotherton, 2015; van Prooijen, 2018) and articles on conspiracy theory beliefs (Shermer,

2014; Oaklander, 2015, Yong, 2008). It is timely to consider formally the strength of the experimental evidence for this claim, especially as it does not follow straightforwardly from compensatory control theory.

Control and conspiracy beliefs

An argument can be made that conspiracy theory beliefs are not, in principle, well suited to function as a compensatory control mechanism. First, conspiracy theories, by accounting for errant data, connecting seemingly unconnected events, and leaving nothing unexplained (Clarke, 2002; Keeley, 1999) provide too much order and structure. According to compensatory control theory people seek an optimal sense of control, not a maximum, and

Kay and Sullivan (2013) suggest that too much structure is just as aversive as too little.

Second, because the conspirators are malevolent by definition (van Prooijen, 2018) and belief in conspiracy theories is accompanied by social stigma (Lantian, Muller, Nurra,

Klein, Berjot & Pantazi, 2018), they are neither “socially acceptable” nor “culturally accessible” – two important criteria for an effective compensatory control mechanism (Kay

& Sullivan, 2013). The reliance on conspiracy theory beliefs is particularly dubious given the multitude of alternative, benevolent, non-stigmatizing, proven substitutes for personal control, such as belief in social progress or in competent authorities (Shepherd, et al., 2011).

Even if these alternatives are not all equally effective at establishing a sense of control, conspiracy theories would seem to be an option of last resort.

152

The current study

Although the claim that conspiracy theories compensate for threatened control has dubious theoretical validity, does it have empirical support? In the current study, we review and meta-analyze the currently available evidence. In addition, we examine potential moderators such as operationalization of conspiracy beliefs, type of experimental manipulation, presence or absence of manipulation check, comparison group, and participant sample. We also examine publication bias by comparing the overall effect of published and unpublished studies.

Method

Literature search and selection criteria.

We conducted our literature search in February 2019, and updated it in June 2019. In line with the recommendations of Card (2011), we searched both discipline-specific databases (PsychInfo) and larger, cross-disciplinary databases (Web of Science, Scopus), using the following keywords: (conspir* OR (conspir* AND (ideation OR belief*

OR theory))) AND (control OR power*). To locate unpublished papers, we searched

ProQuest Dissertations and Thesis (using the same search keywords), Open Access Thesis and Dissertations ( keywords: (conspir* AND control) AND (ideation OR belief OR theory)), and sent e-mails inquiring about any unpublished data to authors who have previously published on this topic. Finally, a call for unpublished data was sent to the list serv of the

Society for Personality and Social Psychology, the Society of Experimental Social

Psychology, and the Society for the Psychological Study of Social Issues.

153

The search yielded 2360 unique records. Figure 1. presents a PRISMA flowchart of the selection process (Moher, Liberati, Tetzlaff, Altman, & The PRISMA Group, 2009). The first author screened the abstracts of these records to find studies that both manipulated participants’ feelings of control and also measured conspiracy beliefs, after which 25 items remained for full text consideration. Of those, studies were included if they both represented original research (i.e. were not previously reported), and whose manipulation check, when reported, was successful.

Based on these criteria we included 15 reports that contained 23 studies, and 45 effect sizes (16 published, 29 unpublished) for meta-analysis.

154

Records identified through database searching (n=3048) and other sources (n = 5)

Identification

Records excluded in the

Records after duplicates title and abstract stage removed (n = 2334) (n = 2360)

Screening

Records excluded at the

Records assessed for full text reading stage: eligibility did not (successfully) (n = 26) manipulate control n = 5; did not report original data Eligibility n = 6 )

Records included in meta- analysis (n = 15) Included

Figure 1. PRISMA-style flowchart showing selection of studies for meta-analysis on conspiracy theories as compensatory control literature.

155

Possible moderators.

Type of experimental manipulation. Control has been experimentally manipulated in a variety of ways, such as recall priming, in which participants recall and describe an event in which they felt out of control (e.g. Whitson & Galinsky, 2008; van Elk & Lodder, 2018), biased questionnaires, in which participants indicate agreement with statements designed to threaten their control (e.g. Sullivan, Landau & Rothschild, 2010), and concept identification, in which participants receive noncontingent feedback (e.g. Imhoff, 2015, Study 3). It is possible that some of these tasks are more effective at eliciting the desired state (i.e. lack of control) than others; consequently, if the hypothesized effect of lack of control on conspiracy beliefs exists, it may be moderated by the type of manipulation used.

Operationalization of conspiracy beliefs. Conspiracy beliefs have also been measured in a variety of ways, for example via judgments of coincidences (Whitson &

Galinsky, 2008), ratings on the Conspiracy Mentality Questionnaire or variants of it (van Elk

& Lodder, 2018; Imhoff, 2015 Study 1), or by asking about the likelihood of a conspiracy in an ambiguous scenario (Whitson, Kim, Wang, Menon & Webster, 2018). These measures can be more abstractly classed as specific (i.e. having to do with particular events or groups) or generic (the tendency to think in conspiratorial terms). Imhoff (2015), for example, implies that lack of control should only increase belief in specific conspiracies, but not the general tendency to believe in conspiracy, which is more or less stable. Thus, the way conspiracy beliefs are measured may be source of variation in effect sizes. We examine this possibility by looking at operationalization of the dependent variable as a potential moderator.

Comparison group. The condition to which threatened control is compared could likewise account for variance in results. For example, some studies use control affirmation

156

(asking participants to recall an event in which they were in control) as a comparison (e.g.

Whitson & Galinsky, 2008), while others use a baseline group (no reference to control; e.g.

Hart & Graether, 2018 study 2). Indeed, all studies that support the compensatory control hypothesis appear to have compared the low control to control affirmation group. To formally test possible differences between these research designs, we include comparison group as a moderator in our analysis.

Manipulation Check. Another, nonobvious source of variability may be the use of a manipulation check. Some researchers have charged that manipulation checks in experimental designs in general are unwise because, among other things, they may undo the effect of the manipulation (Hauser, Ellsworth & Gonzales, 2018). In the case of compensatory control theory specifically, it may be that the opportunity to assert control via self-report is sufficient to undo the (presumably weak) threat posed by the experimental manipulation, which might account for some null findings.

Sample. Participants in majority of the studies were either students or Mechanical

Turk (MTurk) workers. To assess if the effect is more or less strong depending on the population the sample was drawn from, we included sample as potential moderator.

Publication status. Publication bias may lead to larger estimates of an effect size

(Rothstein, Sutton, & Borenstein, 2006). As 64% of the effect sizes were drawn from unpublished studies, we used publication status as a potential moderator. Along with other tests of publication bias, this allowed us to examine whether the published literature in this area systematically differs from the unpublished studies.

157

Analysis strategy.

The title, publication year, authors, sample, dependent measure operationalization, type of experimental manipulation, presence or absence of manipulation check, comparison group, publication status, and effect sizes were extracted from the studies that met the inclusion criteria. When group sample size was not reported, and the authors did not respond to our inquires, equal sample size was assumed across experimental groups. Cohen’s d

(Cohen, 1988), based on information reported in the articles (or, if necessary, on communication with the authors themselves) was obtained with the function escalc from the metafor package (Viechtbauer, 2010); negative values indicate that decreases in control increased conspiracy beliefs, in support of the compensatory control hypothesis. The first author along with two trained research assistants coded the moderators in the studies. The three coders agreed 87% of the time, with disagreements resolved through discussion. Table

1. lists the moderator coding criteria.

Note that, where relevant, separate effect sizes were calculated for each dependent variable and/or each comparison group within each reported study, which introduced some dependency into the data. This dependency was dealt with by using meta-analysis with robust variance estimates (RVE), which adjusts for dependent effect sizes (Hedges, Tipton, &

Johnson, 2010). The RVE method is implemented in R (R Core Team, 2018) in the robumeta package (Fisher & Tipton, 2015), which is suitable for meta-analyses with small numbers of studies (Tipton, 2015). Since it has been recommended to choose weighting based on the predominant dependency (Tanner-Smith & Tipton, 2014), we used the correlated weighting scheme (Hedges et al., 2010), as most of the dependency arose due to studies providing multiple effect sizes (66%) as opposed to studies coming from a single lab (33%).

158

Table 1. Moderator Coding Criteria and number of studies meeting each criteria

Moderator Level Criteria Sample Student The sample consisted of undergraduate students, or predominantly of students. (n = 8 studies)

MTurk The sample consisted of MTurk participants (n = 10 studies)

Other The sample could not be classified into these two categories (n = 5 studies)

Experimental Recall task Participants were asked to recall and describe a time when they felt out of manipulation control. (n = 18 studies)

Other The experimental manipulation was other than the recall task. (n = 5 studies)

Dependent Specific Conspiracy beliefs were operationalized as belief in a specific conspiracy or measure set of conspiracies (e.g. Nasa faked the moon landing, Jews secretly control the banking system). (n = 16 studies)

Generic Conspiracy beliefs were operationalized as generic beliefs, e.g measured with the Conspiracy Mentality Questionnaire (Bruder et al, 2013) or the Generic Conspiracist Beliefs Scale (Brotherton et al., 2013) (n = 9 studies)

Comparison Control The comparison group was a control affirmation group (e.g. a group asked to group affirmation recall a time when they were in control, or in other way affirmed their sense of control) (n = 14 studies)

Baseline The comparison group was a baseline group, who either did nothing, or recalled and described a memory on neutral topic such as watching TV or eating dinner. (n = 14 studies)

Manipulation Yes The authors report using manipulation check. (n = 4 studies) Check No The authors do not report using a manipulation check. (n = 29 studies)

Publication Published The study is published in a peer-reviewed journal. (n = 9 studies) Status Unpublished The study is from a dissertation, unpublished data, manuscript in preparation/in review/in press, books, Online Supplementary Materials etc. (n = 15)

159

Results

Overview of studies

The average experiment had 375 participants (total N=8618 participants; 54% female). In the majority of the experiments, lack of control was elicited with the recall task

(78% of studies). In this task participants are asked to recall and describe an event when they felt completely out of control. The low control group was compared with a control affirmation group in 60% of studies, and with a baseline in 60%7 of studies. Conspiracy beliefs were operationalized as belief in a specific conspiracy theory or theories in 70% of studies, and as a generic attitude in 39%8. A summary of the coded features of all studies included in the meta-analysis, including effect sizes, is given in Table 2.

7 Some of the studies had multiple comparison groups, hence percentages do not add to 100. 8 Some of the studies had multiple dependent variables, hence the percentages do not add to 100. 160

Table 2. Study description and effect sizes

Comparison Experimental Manipulation Publication % Study DV Sample N d group Manipulation check status woman Whitson & Galinsky Control Specific Recall task No Students Published 76% 25 -0.85 (2008), Study 4 affirmation Whitson & Galinsky (2008), Study 6, self Specific Baseline Recall task No Students Published 62% 50 0.02 affirmation Whitson & Galinsky , (2008), Study 6, no Specific Baseline Recall task No Students Published 62% 50 -0.55 affirmation Whitson and Galinsky, Control footnote 35 study, other Specific Recall task No Other Unpublished 46% 41 -0.72 affirmation focused Whitson and Galinsky, Control footnote 35 study, self Specific Recall task No Other Unpublished 46% 41 -0.61 affirmation focused Whitson, thesis, Control Specific Recall task No Students Unpublished 62% 50 0.28 Study 1 affirmation Sullivan et al. (2010), Control Study 2, opponent Specific Other Yes Students Published 54% 59 -0.46 affirmation conspiracies Sullivan et al. (2010), Control Study 2, non-opponent Specific Other Yes Students Published 54% 59 0.20 affirmation conspiracies van Prooijen & Acker Specific Baseline Recall task No Students Published 67% 119 -0.26 (2015), Study 1 van Prooijen & Acker Control Specific Recall task No Students Published 67% 119 -0.73 (2015), Study 1 affirmation Hart & Graether Control Generic Recall task No Mturk Published 46% 422 0.11 (2018), Study 1 affirmation

161

Hart & Graether Generic Baseline Recall task No Mturk Published 49% 831 0.09 (2018), Study 2 Whitson et al. (2018), Control Study 3, promotion Specific Recall task No Students Published 51% 215 -0.49 affirmation focus Whitson et al. (2018), Control Study 3, prevention Specific Recall task No Students Published 51% 215 0.11 affirmation focus Whitson et al. (2018), Study 3 from Supplementary Specific Baseline Recall task No MTurk Unpublished 47% 251 -0.36 materials, promotion focus Whitson et al. (2018), Study 3 from Supplementary Specific Baseline Recall task No MTurk Unpublished 47% 251 0.32 materials, prevention focus Nyhan & Zeitzoff Control Specific Recall task Yes Other Published 45% 2015 -0.03 (2018a), Study 1 affirmation Nyhan & Zeitzoff (2018b), Study 1, Specific Baseline Recall task No Other Published 52% 2170 -0.15 without amendment Nyhan & Zeitzoff Control (2018b), Study 1, Specific Recall task No Other Published 52% 2170 -0.13 affirmation without amendment Nyhan & Zeitzoff (2018b), Study 1, with Specific Baseline Recall task No Other Published 52% 2170 0.10 amendment Nyhan & Zeitzoff Control (2018b), Study 1, with Specific Recall task No Other Published 52% 2170 0.05 affirmation amendment

162

Imhoff Unpublished Generic Baseline Other No Students Unpublished - 44 -0.20 (2015), Study 1 Imhoff Unpublished Control Generic Other No Students Unpublished - 72 0.20 (2015), Study 3 affirmation Lamberty, unpublished Generic Baseline Recall task No Mturk Unpublished 47% 199 -0.11 data Lamberty, unpublished Specific Baseline Recall task No Mturk Unpublished 47% 199 -0.12 data Bukowski, Control Specific Other No Other Unpublished - 120 -0.49 unpublished, Story 1 affirmation Bukowski, Control Specific Other No Other Unpublished - 120 0.11 unpublished, Story 2 affirmation Bukowski, Control Specific Other No Other Unpublished - 120 -0.32 unpublished, Story 3 affirmation Kofta et al., Generic Baseline Other No Other Unpublished 41% 172 -0.19 unpublished, Generic Kofta et al., Specific Baseline Other No Other Unpublished 41% 172 -0.55 unpublished, Jews Kofta et al., Specific Baseline Other No Other Unpublished 41% 172 -0.41 unpublished, Germans Kofta et al., Specific Baseline Other No Other Unpublished 41% 172 -0.35 unpublished, Russians Marques & Natoli, Specific Baseline Recall task Yes Mturk Unpublished 52% 323 -0.22 unpublished Marques & Natoli, Control Specific Recall task Yes Mturk Unpublished 52% 323 -0.07 unpublished affirmation Kumareswaran, thesis, Control Specific Recall task No Students Unpublished 53% 83 0.16 Study 1 affirmation Stojanov et al. Generic Baseline Recall task Yes Mturk Unpublished 65% 194 0.27 unpublished, Study 1 Stojanov et al., Control Generic Recall task Yes Mturk Unpublished 65% 194 0.14 unpublished, Study 1 affirmation

163

Stojanov et al., Generic Baseline Recall task No Mturk Unpublished 61% 197 -0.14 unpublished, Study 3 Stojanov et al., Specific Baseline Recall task No Mturk Unpublished 63% 199 0.02 unpublished, Study 4 Stojanov et al., unpublished, Study 5, Specific Baseline Recall task No Mturk Unpublished 64% 291 -0.03 control related CT Stojanov et al., Control unpublished, Study 5, Specific Recall task No Mturk Unpublished 64% 291 0.14 affirmation control related CT Stojanov et al., unpublished, Study 5, Specific Baseline Recall task No Mturk Unpublished 64% 291 0.01 control unrelated CT Stojanov et al., Control unpublished, Study 5, Specific Recall task No Mturk Unpublished 64% 291 0.02 affirmation control unrelated CT Stojanov et al., unpublished, Study 6, Generic Baseline Recall task No Mturk Unpublished 62% 501 0.14 competent government scenario Stojanov et al., unpublished, Study 6, Generic Baseline Recall task No Mturk Unpublished 62% 501 -0.07 incompetent government scenario

164

Main analysis

We used the robumeta package (Fisher & Tipton, 2015) to fit an intercept-only random effects model. The intercept can be interpreted as the precision-weighted, correlated- effect dependencies adjusted, overall effect size. The same approached was used to calculate the effect sizes for each level of our moderators of interest. It should be noted, though, that many observations are needed to achieve high power in moderation analysis (Hedges &

Pigott, 2004), thus non-significant interaction findings should be interpreted with caution.

We also checked for influential cases with the function influence.measures; no influential cases were detected.

Overall effect size. Forty-five effect sizes were used to compute the overall effect

(see Figure 2.). Of these, descriptively, 26 were in the direction hypothesized by compensatory control theory and 19 in the opposite direction. Effect sizes ranged from -0.85 to 0.32 (see Figure 2.). The overall effect size was d = -0.05 , 95% CI [ -0.11, 0.02], t (16.3)

= -1.45, p = 0.16. The statistic for the ratio of true heterogeneity to total variance (Higgins,

Thompson, Deeks, & Altman 2003) (I2 = 43.13) suggested moderate heterogeneity of the data, justifying further tests of potential moderators. Table 3. gives the effect size estimates for each level of each moderator.

165

Table 3. Estimate of effect size for each level of each moderators, along with significance test of moderation effects

Moderator Level s k d β F 95% CI p

Sample 38 1.22 0.37

Students 8 13 -0.13 [-0.35, 0.09] 0.19

MTurk 10 18 0.02 [-0.06, 0.10] 0.62

Other 5 14 -0.12 [-0.39, 0.13] 0.21

DV -0.14 [-0.26, -0.02] 0.02* operationalization

Specific 16 34 -0.11 [-0.20, -0.02] 0.02*

Generic 9 11 0.03 [-0.07, 0.43] 0.48

Comparison group 0.01 [-0.10, 0.13] 0.82

Baseline 14 23 -0.05 [-0.14, 0.04] 0.27

Control 14 22 -0.05 [-0.16, 0.06] 0.34

Affirmation

Experimental 0.13 [-0.16, 0.43] 0.29 manipulation

Recall task 18 34 -0.03 [-0.09, 0.04] 0.38

Other 5 11 -0.16 [-0.46, 0.14] 0.21

Manipulation check 0.04 [-0.17, 0.26] 0.62

Present 4 7 -0.02 [-0.26, 0.23] 0.8

Absent 19 38 -0.06 [-0.14, 0.02] 0.15

Publication Status 0.01 [-0.14, 0.15] 0.92

Unpublished 15 29 -0.04 [-0.13, 0.05] 0.36

Published 9 16 -0.07 [-0.22, 0.08] 0.29

166

Note: k = number of effect size estimates; s = number of studies; d = Cohen’s standardized difference; β = coefficients from separate meta-regressions where a two level moderator was dummy coded and entered as a predictor; F = Approximate Hotelling – Zhang with small sample correction omnibus test for moderators with more than two levels; 95% CI refer either to β coefficients or d values; p corresponds to β or F. * = significant at < 0.05, + = significant at < 0.1

167

Forest Plot

Studies Effect Size

1. Whitson & Galinsky, 2008 Study 4 −0.848 Whitson doctoral dissertation 0.283

2. Whitson & Galinsky, 2008 Study 6, self affirmation 0.020 Study 6, no affirmation −0.551

3. Whitson & Galinsky, 2008 Study in footntote 35, other focused −0.722 Study in footntote 35, self focused −0.606

4. Sullivan et al., 2010 Study 2, oponent conspiracy −0.459 Study 2, non oponent conspiracy 0.203

5. van Prooijen & Acker, 2015 Study 1, compared to baseline −0.262 Study 1, compared to control affir mation −0.729

6. Hart & Graether, 2018 Study 1, compared to control affir mation 0.112

7. Hart & Graether, 2018 Study 2, comapred to baseline 0.092

8. Whitson et al., 2018 Study 3, promotion focus −0.486 Study 3, prevention focus 0.111

9. Whitson et al., 2018 Study 3 in supplementar y materials, promotion focus −0.357 Study 3 in supplementar y materials, prevention focus 0.324

10. Nyhan & Zeitzoff, The Jouranl of Politics, 2018 Study 1 −0.029

11. Nyhan & Zeitzoff, Political Psychology, 2018 Study 1, compared to baseline, without amendment −0.154 Study 1, compared to control affir mation, without amendment −0.130 Study 1, compared to baseline, with amendment 0.098 Study 1, compared to control affir mation, with amendment 0.049

12. Imhoff, 2015 Study 1 −0.199

13. Imhoff, 2015 Study 3 0.205

14. Lamberty, unpublished Study 1, Generic −0.114 Study 1, Specific −0.117

15. Bukowski, unpublished Study 1, Story 1 −0.487 Study 1, Story 2 0.114 Study 1, Story 3 −0.319

16. Kofta et al., under review Study 3, Generic −0.193 Study 3, Jews −0.553 Study 3, Germans −0.411 Study 3, Russians −0.347

17. Marques & Natoli, unpublished Study 1, comapred to baseline −0.218 Study 1, compared to control affir mation −0.074

18. Kumareswaran, 2014 Study 1 0.164

19. Stojanov et al., under review Study 1, compared to baseline 0.267 Study 1, compared to control affir mation 0.141

20. Stojanov et al., under review Study 3 −0.144

21. Stojanov et al., under review Study 4 0.025

22. Stojanov et al., under review Study 5, compared to baseline, control related conspiracies −0.028 Study 5, comapred to control affir mation, control related conspiracies 0.139 Study 5, compared to baseline, control unrelated conspiracies 0.013 Study 5, comapred to control affir mation, control unrelated conspir acies 0.024

23. Stojanov et al., under review Study 6, competent government scenario 0.137 Study 6, incompetent government scenario −0.072

−2 −1 0 1 2 Effect Size

Figure 2. Forest plot of the effect size for the studies in the meta-analysis. Filled squares indicate effect sizes of the respective studies. The white square indicates the overall effect size.

168

Moderator analysis.

Moderators with three levels.

Sample. Because the variable sample had more than two levels, we performed an omnibus test, following the recommendations by Tanner-Smith, Tipton, and Polanin (2016).

We used the clubSandwhich package (Pustejovsky, 2016) to obtain F value from the

Approximate Hotelling-Zhang test, which indicated that there is no reliable difference among the levels of this variable, F = 1.22, df = 4.77, p = 0.37.

Moderators with two levels.

For the moderators with two levels, we ran separate analyses on effect size using each potential moderator as an independent variable in turn. These analyses revealed a significant effect of DV operationalization, β = - 0.14, 95% CI [-0.26, -0.02], p = 0.02, such that the threats to control increased beliefs in specific conspiracy theories ( d = -0.11, 95% CI [-0.20,

-0.02], p = 0.02), but not generic conspiracy theory beliefs (d = 0.03, 95% CI [-0.07, 0.14], p

= 0.48). Effect sizes did not differ as a function of the presence of a manipulation check, the comparison group used, or how control was manipulated.

Post hoc analysis. Given the significant effect of control beliefs on specific conspiracy theory beliefs, we examined, for exploratory purposes, if this effect is moderated by any of the other variables. None did, though it is worth noting that the effect was significant for studies that omitted manipulation checks (β =-0.13, 95% CI [-0.26, -0.01], p

=0.03), but not for studies that used them (β =-0.06, 95% CI [-0.43, 0.31], p=0.37).

169

Publication bias. We tested for publication bias in two ways. First, we analysed the magnitude of effect sizes as a function of publication status; there was no difference between published and unpublished studies. Second, we aggregated the effect size for the published studies and estimated the adjusted effect size using the trim and fill procedure (Duval &

Tweedie, 2000). In the trim phase of the procedure, the studies that contribute for asymmetrical plot are excluded and an adjusted symmetry effect is calculated. In the fill phase, the excluded studies are replaced along with their “missing” counterparts around the adjusted effect size. The adjusted effect size was estimated to be d = -0.01, 95% CI = [-0.14,

0.12], p = 0.87. As can be seen from the funnel plot in Figure 3., the trim and fill procedure identified three “missing studies”, suggesting publication bias.

Figure 3. Funnel plot obtained with the trim and fill procedure. White circles indicate imputed studies.

170

P-curve analysis. P-curve analysis has been introduced as a method of evaluating sets of statistically significant results for evidential value (Simonsohn, Nelson, & Simmons, 2014).

If the effect in question is true, then the distribution of significant p-values follows a distinct pattern of greater density of low values than high values (i.e., a right skew). On the other hand, if the effect is non-existent (type I error), the distribution of the p-values should be uniform as there is equal probability to obtain any p-value within the null distribution (Hung,

O'Neill, Bauer, & Kohne, 1997). Thus, the shape of the p-curve can help distinguish a set of significant findings that are likely to be replicated (right skew) or not (flat line).

The test has two parts. In the first part the set of p-values is tested for right skew: the probability of obtaining the p-value if the null hypothesis is true is calculated for each p- value. These probabilities are then aggregated to yield a Z test of skew; significance of the Z test suggests a significant right skewed p-curve. In the second part, it is tested if the p-curve is significantly flatter than one would expect if the studies were powered at 33%; if it is, then the set of findings lacks evidential value: if an effect exists, it is negligible. If the distribution of p-values is not flatter than one would expect if the studies were powered at 33% than the p-curve is inconclusive, and more p-values are needed to determine whether the studies contain evidential value.

As the p-curve shape is tied to the probability that a study will yield a significant result, p-curve can also be used to estimate the power of the studies (Simonsohn, Nelson &

Simmons, 2014).

171

Figure 4. P-curve analysis results

We subjected the studies from the meta-analysis to p-curve analysis (see Figure 4.).

The p-curve disclosure table can be found in the Supplementary Materials. Testing for right skew, Z=0.34, p=0.63, did not suggest evidential value; the 33% power test was Z=-1.93, p=0.03, suggesting that direct replications of the examined studies are not expected to succeed. The average power of the studies was 5% which is what would be expected when the true effect size is zero and the significance level is 0.05 (Simmons & Simonsohn, 2017).

The 90% CI around the power estimate suggested that if the same studies were conducted

172

again it is unlikely that more than 25% would replicate (Simmons & Simonsohn, 2017) and the best estimate is that only 5% would be yield significant results.

Discussion

The current study is the first attempt to systematically review and evaluate the available evidence for conspiracy beliefs as a compensatory control mechanism. The results of our meta-analysis offered very little support for that hypothesis: the overall effect size was

-.05, not different from chance, and was not strongly moderated by any of several obvious contenders. The most promising moderator was the nature of the dependent variable, with a significant effect detected for specific theories, but not for generic conspiracy beliefs. We can only speculate about the cause of this difference, but one possibility is that generic conspiracy belief reflects a relatively stable personality trait that is insensitive to momentary threats to control. Another is that generic beliefs are too broad for compensatory control purposes, which might require a more specific wrongdoer as a focal point.

Looking more closely at specific conspiracies only, for which the effect was significant, the effect appeared to be more likely observed if the design did not use manipulation check (though the effect was not significantly different from the effect coming from studies that did use manipulation check). Future studies should explore this trend further; if robust it could prove both methodologically and theoretically important. It is worth nothing, though, that even the effect observed for specific conspiracy theories is weak, suggesting a limited role for lack of control. Moreover, the p-curve analysis did not demonstrate right skew, suggesting that the significant findings are most likely due to chance.

Overall then, the prospects for finding a significant experimental effect of control on conspiracy theory beliefs are not promising.

173

Although the evidence from existing research offers very limited support for the compensatory control hypothesis, it is too soon to reject it. Many experimental manipulations of control have dubious validity. The most common experimental technique – asking participants to recall an uncontrollable situation – is transparent and vulnerable to experimenter demand (Orne, 2009). Even if a change in perceived control is genuine in this paradigm, it is in principle unstable: if a participant’s sense of control can be altered merely by imagining an uncontrollable situation, it should be rectified just as easily, such as by spontaneously recruiting additional, compensatory cognitions. Indeed, some researchers have proposed that simply asking participants about their feelings of control (e.g., in a manipulation check) is sufficient to undo any threat their previous recall produced (Hauser et al., 2018), a trend evident in our data as well. Even at their best, laboratory manipulations that require participants to imagine past threats are likely to pale in comparison to the situations that actually caused the threats. To our knowledge there is no study of such “natural” threats to control, and such a study would be an important complement to the experimental literature.

It is important to acknowledge the limitations of the current analyses as well.

Although we made every attempt to locate all relevant experimental data, it is possible that we did not identify all the existing records. Furthermore, although the number of studies in the analysis is well within the norm (Tipton, 2015, cites a finding that 50% of meta-analysis in the social sciences include fewer than 32 studies), it is still a relatively small absolute number, and conclusions about moderating variables in particular are limited.

To conclude, this is the first study to systematically examine the available evidence for the thesis that lack of control leads to conspiracy beliefs. The current literature, both published and unpublished, offers very limited support for the hypothesis that conspiracy theory beliefs are a compensatory control mechanism, with the most promising moderator being the nature of the dependent measure: there was a significant (albeit weak) effect of

174

control manipulation only when conspiracy beliefs were operationalized in terms of specific conspiracies. If conspiracy beliefs are indeed in part a result of compensatory control processes, the experimental studies so far have not convincingly demonstrated it.

175

References

Bilewicz, M. , Winiewski, M. , Kofta, M. and Wójcik, A. (2013), Harmful Ideas. Political

Psychology, 34: 821-839. doi:10.1111/pops.12024

Brotherton, R. (2015). Suspicious minds: Why we believe conspiracy theories. London,

England: Bloomsbury.

*Bukowski, M. (personal communication, April 27, 2019).

Card, N. A. (2011). Applied meta-analysis for social science research. Retrieved from

https://ebookcentral-proquest-com.ezproxy.otago.ac.nz

Clarke, S. (2002). Conspiracy Theories and Conspiracy Theorizing. Philosophy of the Social

Sciences, 32(2), 131–150. https://doi.org/10.1177/004931032002001

Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences. New York, NY:

Routledge Academic

Collins, N. (2018). Conspiracy theories are more rampant than ever. Can they be stopped?

Accessed on 12/08/2019: https://www.theguardian.com/us-

news/2018/aug/14/conspiracy-theories-misconceptions-rampant-politics

Douglas, K. M., & Sutton, R. M. (2015). Climate change: Why the conspiracy theories are

dangerous. Bulletin of the Atomic Scientists, 71(2), 98-106.

doi:10.1177/0096340215571908

Douglas, K. M., & Sutton, R. M. (2018). Why conspiracy theories matter: A social

psychological analysis. European Review of Social Psychology, 29(1), 256-298.

doi:10.1080/10463283.2018.1537428

Douglas, K. M., Sutton, R. M., & Cichocka, A. (2017). The Psychology of Conspiracy

Theories. Current Directions in Psychological Science, 26(6), 538–542.

https://doi.org/10.1177/0963721417718261

176

Duval, S., & Tweedie, R. (2000). A nonparametric “trim and fill” method of accounting for

publication bias in meta-analysis. Journal of the American Statistical Association, 95,

89-98.

Fisher, Z., & Tipton, E. (2015). robumeta: An R-package for robust variance estimation in

meta-analysis. Retrieved from arXiv preprint arXiv:1503.02220

Fisher, Z., & Tipton, E. (2015). Robumeta: An R-package for robust variance estimation in

meta-analysis. Retrieved from: https://arxiv.org/abs/1503.02220)Hauser, D.J.,

Ellsworth, P.C., & Gonzales, R. (2018). Are Manipulation Checks Neccesary?

Frontiers in Psychology, 9, 998.

Goode, C., Keefer, L. A., & Molina, L. E. (2014). A Compensatory Control Account of

Meritocracy. Journal of Social and Political Psychology, 2, 313-334.

*Hart, J., & Graether, M. (2018). Something's going on here: Psychological predictors of

belief in conspiracy theories. Journal of Individual Differences, 39(4), 229-237.

doi:10.1027/1614-0001/a000268

Hauser, D.J., Ellsworth, P.C., & Gonzalez, R. (2018). Are Manipulation Checks

Necessary? Frontiers in Psychology, 9:998. doi: 10.3389/fpsyg.2018.00998

Hedges, L. V., & Pigott, T. D. (2004). The Power of Statistical Tests for Moderators in Meta-

Analysis. Psychological Methods, 9(4), 426-445. http://dx.doi.org/10.1037/1082-

989X.9.4.426

Hedges, L. V., Tipton, E. and Johnson, M. C. (2010), Robust variance estimation in meta‐

regression with dependent effect size estimates. Research Syntesis. Methods 1, 39-65.

doi:10.1002/jrsm.5

Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring

inconsistency in meta-analyses. British Medical Journal, 327(7414), 557-560.

177

Hung, H. M. J., O'Neill, R. T., Bauer, P., & Kohne, K. (1997). The behavior of the p-value

when the alternative hypothesis is true. Biometrics, 11-22. doi:10.2307/2533093

*Imhoff, R. (2015). Beyond (right-wing) authoritarianism: Conspiracy mentality as an

incremental predictor of prejudice. In M. Bilewicz, A. Cichocka & W. Soral

(Eds.) The Psychology of Conspiracy Theories (pp.122-141). East Sussex: Routledge.

Jolley, D., Douglas, K. M., Leite, A. C., & Schrader, T. (2019). Belief in conspiracy theories

and intentions to engage in everyday crime. British Journal of Social

Psychology, doi:10.1111/bjso.12311

Jolley, D., Meleady, R., & Douglas, K. M. (2019). Exposure to intergroup conspiracy

theories promotes prejudice which spreads across groups. British Journal of

Psychology, doi:10.1111/bjop.12385

Kay, A. C., Gaucher, D., Napier, J. L., Callan, M. J., & Laurin, K. (2008). God and the

government: Testing a compensatory control mechanism for the support of external

systems. Journal of Personality and Social Psychology, 95(1), 18-35.

http://dx.doi.org/10.1037/0022-3514.95.1.18

Kay, A.C. & Sullivan, D. (2013). Cultural unity and diversity in compensatory control

processes. In M.J. Gelfand, C.-y. Chiu, & Y.-y. Hong (Eds.): Advances in Culture and

Psychology (pp. 181-226): Oxford, UK: Oxford University Press.

Kelley-Romano, S. (2008). “Trust No One: The Conspiracy Genre on American Television.”

Southern Communication Journal 73, 105-121.

Keeley, B. (1999). Of conspiracy theories. Journal of Philosophy 96, 109–26.

doi: 10.2307/2564659

*Kofta, M., Bilewicz, M., & Soral, W. (under review). What Breeds Conspiracy

Antisemitism? The Role of Political Powerlessness and Meaninglessness in Belief in

the Jewish Conspiracy.

178

*Kumareswaran, D. J. (2014). The Psychopathological Foundations of Conspiracy theorists.

Wellington: Victoria University of Wellington.

*Lamberty, P. (2019). [ControlExperiment]. Unpublished raw data.

Landau M. J., Kay A. C., Whitson J. A. (2015). Compensatory control and the appeal of a

structured world. Psychological Bulletin, 141, 694–722.

Lantian, A., Muller, D., Nurra, C., Klein, O., Berjot, S., & Pantazi, M. (2018). Stigmatized

beliefs: Conspiracy theories, anticipated negative evaluation of the self, and fear of

social exclusion. European Journal of Social Psychology, 48(7), 939-954.

doi:10.1002/ejsp.2498

Laurin, K., Kay, A. C., & Moscovitch, D. A. (2008). On the belief in God : Towards an

understanding of the emotional substrates of compensatory control. Journal of

Experimental Social Psychology, 44, 1559–1562.

https://doi.org/10.1016/j.jesp.2008.07.007

*Marques, M., & Natoli, E. (2019). [Control_conspMarques.spv]. Unpublished raw data.

Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & the PRISMA group. (2009). Preferred

reporting items for systematic reviews and meta-analyses: The PRISMA

statement. Journal of Clinical Epidemiology, 62(10), 1006-1012.

doi:10.1016/j.jclinepi.2009.06.005

*Nyhan, B,. & Zeitzoff, T. (2018b), Fighting the Past: Perceptions of Control, Historical

Misperceptions, and Corrective Information in the Israeli‐Palestinian Conflict.

Political Psychology, 39, 611-631. doi:10.1111/pops.12449

*Nyhan, B., & Zeitzoff, T. (2018a). Conspiracy and misperception belief in the middle east

and North Africa. Journal of Politics, 80(4), 1400-1404. doi:10.1086/698663

Oaklander, M (2015, August). Here’s why people believe in conspiracy theories. Time.

Retrieved from: http://time.com/3997033/conspiracy-theories/

179

Orne, M.T. (2009). Demand Characteristics and the Concept of Quasi-Controls. In R.

Rosenthal and R.L. Rosnow (Eds.) Artifacts in Behavioural Research: Robert

Rosenthal and Ralph L. Rosnow’s Classic Books. Retrieved from

http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780195385540.001.000

1/acprof-9780195385540-chapter-28

Pichaske, K. (Producer) & Alpert, D. (Producer/Director) (2018). Pulling the Thread. United

States: Kindling Group.

Pustejovsky, J. E. (2016). clubSandwich: Cluster-robust (sandwich) variance estimators with

small-sample corrections (R package version 0.2.1).

URL: https://CRAN.R-project.org/package=clubSandwich

R Core Team (2018). R: A language and environment for statistical computing. R Foundation

for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.

Rothstein, H. R., Sutton, A. J., & Borenstein, M. (2006). Publication bias in meta-analysis:

Prevention, assessment and adjustments. West Sussex, UK: Wiley. Retrieved from

http://www.springerlink.com/index/ L8N2877843431200.pdf

Shermer, M. (2014, December). Conspiracy Central. Scientific American, 311 (6), 94.

Sharma, M., Yadav, K., Yadav, N., & Ferdinand, K. C. (2017). Zika virus pandemic—

analysis of Facebook as a social media health information platform. American Journal

of Infection Control, 45(3), 301–302.

Shepherd, S., Kay, A. C., Landau, M. J., & Keefer, L. A. (2011). Evidence for the specificity

of control motivations in worldview defense : Distinguishing compensatory control

from uncertainty management and terror management processes. Journal of

Experimental Social Psychology, 47, 949–958.

https://doi.org/10.1016/j.jesp.2011.03.026

180

Simmons, J. P., & Simonsohn, U. (2017). Power Posing: P-Curving the Evidence.

Psychological Science, 28(5), 687–693. https://doi.org/10.1177/0956797616658563

Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2014). p-Curve and Effect Size:

Correcting for Publication Bias Using Only Significant Results. Perspectives on

Psychological Science, 9(6), 666–681. https://doi.org/10.1177/1745691614553988

Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2014). P-curve: A key to the file-

drawer. Journal of Experimental Psychology: General, 143(2), 534-547.

Stevens, M. (2019). Falling Trust in Government Makes it Harder to Solve Problems,

Americans Say. Retrieved from:

https://www.nytimes.com/2019/07/22/us/politics/pew-trust-distrust-survey.html on

23/07/2019

*Stojanov, A., Bering, J.M., Halberstadt, J. (under review). Does Perceived Lack of Control

Lead to Conspiracy Theory Beliefs?

*Sullivan, D., Landau, M. J., & Rothschild, Z. K. (2010). An existential function 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(3), 434-449. http://dx.doi.org/10.1037/a0017457

Tanner-Smith, E. E., Tipton, E., & Polanin, J. R. (2016). Handling complex meta-analytic

data structures using robust variance estimates: A tutorial in R. Journal of

Developmental and Life-Course Criminology, 2, 85–112. doi:10.1007/s40865-016-

0026-5

Tanner‐Smith, E. E. and Tipton, E. (2014), Robust variance estimation with dependent effect

sizes: practical considerations including a software tutorial in Stata and SPSS.

Research Synthesis Methods, 5, 13-30. doi:10.1002/jrsm.1091.

181

Tipton, E. (2015). Small sample adjustments for robust variance estimation with meta-

regression. Psychological Methods, 20(3), 375-393.

Tsipursky, G. (2018). (Dis)trust in Science. Can we cure the scourge of misinformation.

Retrieved from https://blogs.scientificamerican.com/observations/dis-trust-in-science/

on 23/07/2019 van Elk, M. & Lodder, P. (2018). Experimental Manipulations of Personal Control do Not

Increase Illusory Pattern Perception. Collabra, 4, 1–22.

*van Prooijen, J. -W., & Acker, M. (2015). The influence of control on belief in conspiracy

theories: Conceptual and applied extensions. Applied Cognitive Psychology, 29(5),

753-761. doi:10.1002/acp.3161 van Prooijen, J.-W. (2018). The psychology of conspiracy theories. London: Routledge.

Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of

Statistical Software, 36(3), 1-48.

*Whitson, J. (2007). Voices in the static: The false perception of patterns in the noise of

everyday life (Order No. 3284140). Available from ProQuest Dissertations & Theses

A&I. (304831516). Retrieved from

https://search.proquest.com/docview/304831516?accountid=14700

*Whitson, J. A., Kim, J., Wang, C. S., Menon, T., & Webster, B. D. (2019). Regulatory

Focus and Conspiratorial Perceptions: The Importance of Personal Control.

Personality and Social Psychology Bulletin, 45(1), 3–15.

https://doi.org/10.1177/0146167218775070

*Whitson, J., & Galinsky, A. D. (2008). Lacking control increases illusory pattern

perception. Science, 322(5898), 115-117. https://doi.org/10.1126/science.1159845

Yong, E. (2008, December). Lacking control drives false conclusions, conspiracy theories

and superstitions. National Geographic. Retrieved from:

182

https://www.nationalgeographic.com/science/phenomena/2008/12/27/lacking-control- drives-false-conclusions-conspiracy-theories-and-superstitions/

183

Chapter 5: Perceived Lack of Control and Conspiracy Theory

Beliefs in the Wake of Political Strife and Natural Disaster

184

Perceived Lack of Control and Conspiracy Theory Beliefs in the Wake of Political Strife

and Natural Disaster

Ana Stojanov

Department of Psychology

University of Otago

Jesse M. Bering

Centre for Science Communication

University of Otago

Jamin Halberstadt

Department of Psychology

University of Otago

Address for correspondence:

Ana Stojanov

William James Building

275 Leith Walk, Dunedin 9016, New Zealand e-mail: [email protected]

Word count: 4793

185

Abstract

While lack of perceived control is plausible explanation of conspiracy beliefs, the experimental evidence so far has provided only mixed support, due in part, we argue, to limits of control manipulations in the laboratory. We present two studies in naturalistic settings that provide evidence that the control hypothesis has merit. In the first study,

Macedonian participants completed a conspiracy ideation scale immediately after a national referendum on the country’s name change from Macedonia to North Macedonia, and one year after. The opposition, whose perceived control was presumably lowered after the name change took place, increased their conspiracy beliefs, but supporters did not. Study 2, conducted with American participants in the wake of a series of devastating tornadoes, conceptually replicated and expanded the first study: the effects were evident only for the threatening event-related conspiracy beliefs. Taken together these studies expand the inconclusive experimental literature and offer the first evidence about the causal link of lack of control for conspiracy beliefs in naturalistic settings.

Keywords: conspiracy beliefs, perceived control, compensatory control, natural disasters, political upheaval

186

The Food and Drug Administration is withholding the cure for cancer. The sky is being sprayed with harmful chemicals. Muslim immigrants are plotting to attack the country from within. These are prototypical examples of conspiracy theories. Although there is currently no agreed-upon definition used by the research community, conspiracy theories are generally thought of as implausible, unwarranted claims that significant events are being caused by malevolent, clandestine groups. Such “theories” typically contradict explanations provided by the relevant epistemic authorities, and tend to be embedded in a broader (often socio-political) explanatory framework (Stojanov & Halberstadt, 2019).

The psychological underpinnings of conspiracy beliefs have become the subject of increased research interest (e.g., Brotherton & French, 2015; Cichocka, Marchlewska & de

Zavala, 2016; Douglas & Sutton, 2011; Imhoff & Lamberty, 2017, 2018; Swami et al., 2011; van Prooijen, 2016; van Prooijen, Douglas & De Inocencio, 2018), as researchers have posited various motivating forces behind people’s endorsement of such beliefs. These include the need for uniqueness (Imhoff & Lamberty, 2017; Lantian, Muller, Nurra & Douglas,

2017), the need to see the world as ordered and structured (Whitson & Galinsky, 2008), and, in particular, the need for personal control (van Prooijen & Acker, 2015). Several studies have demonstrated that when participants’ sense of control is threatened, such as by asking them to recall and describe an event when they lacked control (van Prooijen & Acker, 2015), or by reminding them about everyday uncontrollable hazards (Sullivan, Landau &

Rothschild, 2010), they report stronger conspiracy beliefs relative to control groups.

According to Kay and colleagues’ compensatory control theory, one of the better articulated treatments of the control hypothesis, conspiracy theory beliefs help those who have lost a sense of personal control to restore their worldview that existence is non-random, ordered and structured (see Kay, Gaucher, Napier, Callan & Laurin, 2008; Kay, Sullivan &

Landau, 2015). Belief in nefarious, conspiring agents is not the only way that individuals can

187

regain such a worldview; other restorative “agents” include God (Laurin, Kay, &

Moscovitch, 2008), the government (Shepherd, Kay, Landau, & Keefer, 2011), and even abstract constructs such as meritocracy (Goode, Keefer, & Molina, 2014). However, under the right conditions, explaining unsettling events as the product of coordinated, secretive actors can be psychologically appealing, and presumably a conspiracy theory is preferable to no explanation at all (Marchlewska, Cichocka & Kossowska, 2018).

A closer look at the evidence, however, reveals that the link between personal control and conspiracy beliefs has not been definitively established. For practical as well as ethical reasons, virtually all research on control and conspiracy beliefs is conducted in the laboratory, using various priming techniques to challenge participants’ sense of control, or more often to remind them of times when their sense of control has been challenged. In a typical experiment, one group of participants, usually undergraduate students, is asked to recall and describe an incident in which they felt they did not have any control over the situation, while another group is asked to do the same for an event in which they were in complete control (or for a neutral event with no control implications). They then answer a questionnaire assessing their belief in conspiracy theories, operationalized as either belief in a specific conspiracy claims or as generic conspiracy beliefs (Brotherton, French & Pickering,

2013; Bruder, Haffke, Neave, Nouripanah & Imhoff, 2013; Stojanov & Halberstadt, 2019).

Although many of these studies do suggest that people compensate for diminished personal control by endorsing conspiratorial beliefs (e.g., van Prooijen & Acker, 2015; Whitson &

Galinsky, 2008), other studies (e.g., van Elk & Lodder, 2018) find the opposite, such that participants whose control is threatened report weaker conspiracy beliefs than participants whose control is not threatened. Still other studies report no difference in conspiracy beliefs at all (e.g., Hart & Graether, 2018; Nyhan & Zeitzoff, 2018; Stojanov, Bering & Halberstadt, under review).

188

It is unclear why the evidence for the compensatory control hypothesis is so variable, but a likely component is the relatively weak effects of experimental manipulations of control. The standard experimental technique of asking participants to recall an uncontrollable situation is dubious for several reasons. Such manipulations are transparent to participants and subject to experimenter demand (Orne, 2009). But even assuming that participants truthfully and accurately report their sense of personal control, manipulations based on brief, abstract threats to control are in principle unstable: if a participant’s sense of control can be altered by imagining an uncontrollable situation, it should be able to be rectified just as easily. Some participants may exhibit psychological reactance (Brehm, 1989) and restore their sense of control without waiting for the opportunity to endorse conspiracy theories, potentially overcompensating for the experimental threat in the process. Indeed,

Hauser, Ellsworth and Gonzales (2018) have argued that merely answering a manipulation check, such as having participants report on their sense of control after it has been experientially threatened, is sufficient to undo the effects of the manipulation. Moreover, there are some indications that the recall task may not always be effective in manipulating participants’ feelings of control in the first place (e.g., van Elk & Lodder, 2018).

Even at their best, experimental manipulations in which participants imagine or recall threats to control pale in comparison to the situations that actually caused the threats. To our knowledge there is no study of conspiracy beliefs and perceived control in their “natural” environment. Such an approach would be an important complement to the experimental literature: from a methodological perspective, examining people’s conspiracy beliefs following a “real” threat to their perceived sense of control provides a more powerful and ecologically valid way to address the central claims of the compensatory control model; from a theoretical perspective, this approach would speak directly to the sort of social and political

189

events that appear to motivate conspiracy beliefs in the first place (van Prooijen & Douglas,

2017).

We examine how lack of control affects conspiracy beliefs in real contexts: a political crisis in North Macedonia (Study 1); and a natural disaster in North America (Study 2).

Although the two events could hardly be less similar on the surface, they both represent acute and significant challenges to feelings of control among the individuals who experienced them.

Study 1: North Macedonian Political Crisis, 2018-2019

Background

The state of Yugoslavia disintegrated between 1989 – 1992, and its constituent parts, the Socialist Republic of Bosnia and Herzegovina, the Socialist Republic of Croatia, the

Socialist Republic of Slovenia, and the Socialist Republic of Macedonia, became independent countries9. Of these, we were especially interested in the Socialist Republic of Macedonia, which declared its independence from Yugoslavia in 1991 to create the Republic of

Macedonia. This name was, however, disputed by Greece on the grounds that Macedonia already describes a region in Greece, and that “the Republic of Macedonia” implies territorial claims by this state against the Greek province with the same name. Thus, Greece objected to the country being recognized by its constitutional name (Republic of Macedonia), which in turn resulted in the UN admitting membership to the country under the reference, the Former

Yugoslav Republic of Macedonia (FYROM).

9 The Socialist Republic of Serbia and the Socialist Republic of Montenegro were initially united as the State Union of Serbia and Montenegro, but Montenegro declared independence in 2006.

190

Moreover, in the early 1990s Greece imposed a trade embargo and, until recently, blocked the country’s ascension into EU and NATO. After nearly three decades of conflict, the two countries found a mutually acceptable solution with the signing of the Prespa agreement in June 2018. Among other things, this agreement meant that Republic of

Macedonia would change its name to the Republic of North Macedonia. However, the opposition parties in both countries found the Prespa agreement unacceptable and nationwide protests were initiated. The Greek opposition argued that too many concessions were made in the agreement, such as recognition of Macedonian nationality and language (Smith, 2019); and the Macedonian opposition argued that the name change threatened national identity.

The Prespa agreement needed Macedonian approval via a national referendum.

However, sensing that public opinion was not supportive of the name change, the

Macedonian government determined that the referendum would be consultative rather than obligatory, and framed the question in a way many considered manipulative: “Are you in favour of European Union and NATO membership by accepting the agreement between the

Republic of Macedonia and the Republic of Greece?”. In response, a movement emerged on social media calling on the population to boycott the referendum entirely (#bojkotiram), based on the belief that if the required 50% voter turnout was not met, the name change would not take place. Indeed, voter turnout for the referendum, which took place September

30th of 2018, only reached 37% of the population, and although those who voted were overwhelmingly in favor (94%), it was the opposition who celebrated victory. However, as it turns out, the opposition was wrong: despite the failure to obtain the minimum voter turnout,

Macedonia officially changed to North Macedonia in February 2019 (see Figure 1 for timeline of events).

Thus, the political situation in Macedonia created a natural experiment, in which some citizens’ sense of control, but not others’ was threatened by a salient and emotional political

191

event. To test the effect of this natural manipulation on conspiracy beliefs, we measured those beliefs at two points in time: immediately after the opposition’s apparently successful boycott

(when their sense of control was presumably high); and one year later, after the name change had already been into effect for six months (when their sense of control was low). If lack of control leads to conspiracy beliefs, then the opposition should endorse conspiracy beliefs to a greater extent at the second point than the first, but the opposite should be true for name- change supporters.

Referendum Name changed officially

September February September 2018 2019 2019

Time 1 survey Time 2 survey

Figure 1. Timeline of events between September 2018 and September 2019

192

Method

Participants. A local market research company was hired for recruiting participants. A representative sample of 307 ethnic Macedonians (the ethnic majority, for whom the name change is most relevant) were sent the survey both immediately (Time 1) and one year after

(Time 2) the referendum. For the purposes of the current research, we present data only from participants who reported that they either voted in favor of the name change, or boycotted the referendum/voted no (i.e., excluding those who did not vote for other reasons or did not disclose their behavior), and who provided data at both time points (overall dropout rate was

48%, while the dropout rate only for the participants who voted yes or boycotted/voted no was 37%). This left 116 participants in total, 61 opposers (boycotters or “no” voters”, 30 females, 31 males, mean age was 42.96 years, SD= 15.31) and 55 supporters (26 females, 29 males, mean age M=43.78 years, SD=14.34). The majority had a Bachelor’s degree (50.9%), followed by high school degree (33.6%), Master’s degree (6%) or Associate degree (6%).

Only 1.7% had a doctoral degree and 1.7% had finished only primary education. Supporters and opposers did not differ in terms of their education level, χ2(5)=4.91, p=0.427.

Procedure. A Macedonian market research company, GfK Skopje, was tasked with data collection. Participants were directed to the Qualtrics online platform to complete the survey, which was part of a larger study on an unrelated topic10. All materials were written first in English and then translated independently by two people fluent in English and

Macedonian (one of whom was the first author); minor disagreements in wording were resolved via discussion.

Participants were told that the survey concerned attitudes towards the country’s proposed name change. Participants first answered questions about their emotional state (how

10 The larger survey was looking at identification, identity fusion and the country name as a sacred value.

193

aroused and positive they felt about the name change), interest in the issue of the name change, and their sociocultural identification with Macedonia. Next, they were asked to select which one of five options best reflected their situation: “I voted yes”; “I voted against”; “I boycotted”; “I didn’t vote”; “I’d rather not say”). The option “I didn’t vote” (selected by 51 participants) was offered to distinguish those that did not vote as a political expression from those who did not vote for other reasons. After participants answered the voting question, they were asked to elaborate on their answer in a text box11. No word or time limitations were imposed.

Participants’ conspiracy theory beliefs were operationalized as a score on the

Conspiracy Theory Ideation subscale of the Conspiracy Mentality Scale (Stojanov &

Halberstadt, 2019). This subscale consists of seven items (e.g., “Events on the news may not have actually happened ”; “The alternative explanations for important societal events are closer to the truth than the official story”; “Many so called ‘coincidences’ are in fact clues as to how things really happened”; etc.). Participants rated their agreement with each statement on a 7-point scale anchored at “strongly disagree” and “strongly agree”.

Finally, participants completed a demographic questionnaire, including age, religious affiliation (Christian, Muslim, Atheist, Agnostic, None, Other), gender (male or female), and level of education.

Results and Discussion

11 Qualitative analysis of the answer is part of the larger study on the name change.

194

The descriptive statistics for the dependent variable, by voting behavior (opposers vs supporters) are given in Table 1.

Table 1. Means [and 95% CI] for conspiracy theory ideation (CTI) at time 1 (2018) and time 2 (2019).

Time 1 Time 2

CTI CTI Mean

Supporters 4.62 4.56 4.59

[4.35-4.89] [4.26-4.85] [4.34-4.82]

Opposers 5.19 5.48 5.34

[4.93-5.44] [5.20-5.77] [5.10-5.57]

Mean 4.91 5.02 4.96

[4.72-5.09] [4.82-5.23] [4.79-5.14]

195

Figure 2. Change in conspiracy theory ideation for the supporters and opposers (error bars represent 95% CI)

To test the hypothesis that conspiracy theory beliefs increase after a loss of political power, we conducted a mixed measures ANOVA (voting behavior: support versus oppose) x survey wave: Time 1 vs Time 2), with wave treated as a within subjects factor. There was a main effect of voting behavior, with opposers expressing significantly stronger conspiracy

2 beliefs than supporters, F (1, 114) = 18.412, p < 0.001, ηp = 0.14. There was no significant

2 main effect of wave, F (1, 114) = 1.698, p = 0.195, ηp = 0.015), however, there was a

2 marginally significant interaction with voting behavior, F (1, 114) = 3.949, p = 0.05, ηp =

0.03 (see Figure 2). This remained true even after controlling for change in self-reported

2 positivity and arousal and education, F (1,111) = 3.429, p =0.07, ηp = 0.03. Within-subjects

196

contrasts partially supported our hypothesis: comparing Time 1 and Time 2 scores revealed a significant increase in conspiracy theories for the opposers after the name change took effect,

2 F(1, 58) = 6.613, p = 0.01, ηp = 0. 10, but no difference for supporters, F(1, 52) = 0.076, p =

2 0.78, ηp =0.001

The results offer some support for the idea that conspiracy beliefs increase when control is threatened, and represent the first test of the control hypothesis in a natural setting.

Participants opposing the name change, whose efforts to prevent it were subsequently frustrated, plausibly felt that their personal, political and ingroup power had been compromised. At Time 2, these participants, but not the supporters of the name change, reported stronger conspiratorial thinking, even though they were surveyed six months after the control-threatening event. This is in line with reasoning of other researchers (e.g. Imhoff,

2015) who suggest that it is chronic lack of control that elicits conspiracy beliefs, thus potentially explaining why one-off experimental manipulations may produce weak or

2 inconsistent effects. Indeed, the effect among the control-threatened group (ηp = 0.10, equivalent to d = 0.66) far exceeded the mean effect size obtained in a recent meta-analysis of experimental manipulations of control on conspiracy beliefs we conducted (Stojanov &

Halberstadt, under review) on 45 effect sizes across 23 studies (d = 0.07), hinting at a power- based explanation for previous, uneven experimental findings. This same meta-analysis also revealed a moderating effect of conspiracy belief operationalization, such that control manipulations affected endorsement of specific conspiracy theories more strongly than more abstract conspiracy beliefs (as used in the current studies). In a different paper the authors suggested that conspiracy beliefs that are related to the control threat may be especially likely to be endorsed (Stojanov, Bering & Halberstadt, under review). Thus, effect observed in the current study may have been stronger if measured as endorsement of a specific theory directly related to the political crisis.

197

Contrary to our expectation, supporters of the name change did not show the opposite pattern to the opposers; their conspiracy beliefs remained stable. Although we can only speculate about this null result, it is possible that supporters’ control was not in fact threatened by the ultimate outcome of the referendum. Indeed, both the opposers and the supporters celebrated victory at the time of the referendum – the opposers because the referendum did not reach the census, and the supporters because the overwhelming majority of those who voted found the name change acceptable. By September 2019 it was clear that supporters’ beliefs were justified, but supporters themselves may never have thought otherwise.

Unfortunately, our dataset, which was part of a larger, unrelated study, did not include a direct measure of perceived control, so our mechanistic interpretation of the effects cannot be confirmed empirically. Although we controlled for some possible confounds, such as different educational backgrounds of opposers versus supporters, and differential emotional reactions to the name change, it remains possible that that control-threatened participants’ beliefs changed for other reasons, or indeed that their control was not threatened to begin with.

To address the last concern, we conducted a post hoc survey to see whether people do indeed perceive political upheaval as a threat to personal control. We asked 97 Mechanical

Turk workers (52 males, 45 females, mean age M= 35.77, SD=9.47) what kinds of events tend to influence people’s sense of control. Participants were asked to “think about some of the situations in your life that you have [do not have] control over” (between subjects), and to briefly describe ten such situations (see the Appendix for complete list of controllable and uncontrollable event categories). The first author classified the situations into categories, and a research assistant independently classified 20% of the statements. There agreement rate between the coders was 82%.

198

Politics, arguably the best-fitting category for the Macedonian referendum, ranked 7th

(f = 18), suggesting that while the event we considered is a plausible context in which control might be challenged, there are others that produce more psychologically significant threats.

For example, the third most frequently mentioned factor was the weather (f = 47), and victims of extreme weather events commonly report an acute loss of control (Afifi, Afifi, & Merrill,

2014). In Study 2, we took advantage of this insight and examined the real-time relationship between perceptions of control and conspiracy beliefs immediately following a series of more than 500 tornadoes that struck the American Midwest in May, 2019. In addition to a new participant sample and a different challenge to control, Study 2 added an explicit measure of perceived control, and additional measures of conspiracy theory beliefs. The latter included weather-related conspiracies, allowing us to test the hypothesis that threats to control encourage the endorsement of threat-related conspiracy theories to a greater extent than unrelated or abstract beliefs. Finally, to implicate the tornadoes themselves as the cause of control threat we additionally measured the impact they had on individual participants. We predicted that impact would predict conspiracy theory beliefs (general and/or specific), via their effect on individuals’ sense of control.

Study 2: Midwestern American Tornadoes, 2019

Method

Participants. Two hundredth and sixty-six (114 males, 150 females, 2 “other”)

MTurk workers from Colorado, Indiana, Iowa, Kansas, Missouri, New Jersey, Ohio,

Pennsylvania and Texas took part in the study12 at Time 1. Their average age was 37.74 years

12 Programming of the survey did not allow participants from other States to participate

199

(SD = 13.22, range 18 – 81 years); 41.7% had a Bachelor’s degree, 35.3% high school degree, 14.7% Master’s degree, 2.3% doctoral degree, 0.8% did not have a degree, and 5.3% had an “other” degree. Fifteen participants failed an attention check question and were removed from the final analysis in Time 1. For Time 2 (September) the retention rate was

61%; those who agreed to participate did not differ from those who declined in terms of Time

1 perceived control, affectedness or conspiracy beliefs.

Materials. Impact of tornadoes. To assess the extent to which participants were affected by the tornado we adapted six questions used previously by Segal and colleagues (Segal, Jong, &

Halberstadt, 2018): “To what extent did you suffer physical harm as a result of the tornado?”;

“To what extent did you suffer psychological harm as a result of the tornado?”; “To what extent was your home damaged by the tornado?”; “To what extent was the area where you live damaged by the tornado?”; “To what extent did you feel your life was in danger during the tornado?”; “Overall to what extent has your life been affected by the tornado?”

Participants answered on a 9-point scale anchored at 0 = not at all and 8 = very much so. Answers were combined into one measure by averaging across all the answers.

Conspiracy beliefs. Conspiracy beliefs were assessed in three ways.

Generic conspiracy beliefs. As in Study 1, we used the conspiracy theory ideation subscale of the conspiracy mentality scale (Stojanov & Halberstadt, 2019). Among the items of this scale was an attention check question (“To make sure you read attentively please select strongly agree”).

Specific conspiracy beliefs. To measure specific conspiracy beliefs (as opposed to a general tendency to believe conspiracy theories, as indicated by responses to the foregoing

200

subscale) we used the Conspiracy Theory Beliefs Inventory (Swami et al., 2011), a 15-item scale that measures belief in specific conspiracy theories (e.g., “The Apollo moon landings never happened and were staged in a Hollywood film studio.” Participants answer on a 9- point scale (1 = completely false to 9 = completely true).

Weather-related conspiracy theories. Finally, we measured belief in weather-related conspiracy theories with the following three items: “The adverse weather conditions we are experiencing right now are a result of a weather manipulation experiment”; “The trails left in the sky by planes are evidence of a technology used to change the weather”; “The weather is controlled by multiple governments to help various industries and hurt others.” Participants indicated their agreement with these statements on a 7–point scale (1= strongly disagree to 7

= strongly agree). Cronbach alpha was 0.94 and removing any of the items lowered the coefficient.

Perceived control. Perceived control was measured with Pearlin and Schooler’s

(1978) Mastery Scale. The scale consists of seven items (e.g., “What happens to me in the future mostly depends on me”), five of which are reverse coded. Participants respond on a four-point scale (1= strongly disagree to 4 = strongly agree), and responses are averaged such that higher scores indicate a higher perception of the self being in control.

Procedure. Participants were surveyed on June 1, 2019. After giving informed consent, they were directed to the Qualtrics platform where they read about the purpose of the study (“investigating people’s perceptions about natural disasters and world affairs”). Those who agreed to participate then completed the measure of the tornadoes’ impact, the three conspiracy belief scales in random order, the Mastery Scale, and, finally, a series of

201

demographic questions, before being debriefed. Participants were contacted again in

September 2019 and asked to complete the same set of measures.

Results

Time 1 analysis

Perceptions of control were significantly related to all three measures of conspiracy

beliefs (see Table 2), but the relationship was significantly stronger for weather related

conspiracies than for specific (but weather-unrelated) conspiracy theories, z = -3.573, p <

0.01, or for nonspecific conspiracy ideation, z = -2.317, p = 0.01 (correlations which did not

differ from each other), providing some support for the idea that it is threat-related conspiracy

beliefs that are particularly appealing.

Table 2. Matrix of correlations between the conspiracy beliefs, perceived control and affectedness

Generic Specific Weather related conspiracy conspiracy conspiracy Control Affectedness theories theories theories

Generic CT 1 .74** .59** -.21** .28** Specific CT 1 .65** -.15* .32** Weather related CT 1 -.33** .52**

Control 1 -.20** Affectedness 1

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

202

Path analysis using the lavaan package in R (Rosseel, 2012), with affectedness as independent variable, perceived control as mediator, and all three measures of conspiracy beliefs as dependent variables (see Figure 3), revealed mediating effect of control only for weather related conspiracy beliefs 95%CI [0.012, 0.074], suggesting that control specifically increases belief in threat-related conspiracies (the confidence intervals for the indirect effect for the generic, 95%CI [0.000,0.051], and specific conspiracy beliefs 95% CI [-0.009,0.048] crossed zero).

Figure 3. Path analysis from affectedness to conspiracy beliefs via control. Note: ** p < 0.01, * p < 0.05. Regression weights represent standardized coefficients

203

Time 2 analysis

To determine the persistence of conspiracy beliefs once the immediate threat had subsided, we contacted participants three months later (i.e. September 2019). As expected, feelings of control were higher three months after the tornadoes (see Table 3). A 2 (time:

September versus June) x 3 (Conspiracy beliefs type: Generic, Specific, or Weather-related) repeated measures ANOVA revealed a main effect of conspiracy beliefs, F (2, 302) =

192.910, p < 0.0001, f=1.13, such that weather related conspiracy beliefs were endorsed less strongly then either abstract beliefs, F (1,151) = 288.866, p < 0.001, f=1.38, or specific beliefs. F (1,151) = 237.099, p < 0.001, f =1.25. Importantly, there was a marginal time x conspiracy belief interaction, F (2, 203) = 2.832, p = 0.06, f=0.13, such that weather-related conspiracy beliefs were higher, but generic and specific conspiracy beliefs were lower, immediately after the tornadoes than three months later, although none of the three measures differed significantly over time (see Table 3).

Although conspiracy beliefs were stable on average, individuals’ beliefs over time varied. To determine whether that change was related to changes in perceived control over the same period, we computed change scores for all variables (subtracting the September score from the June score) and regressed all three measures of conspiracy beliefs simultaneously on control. The analysis revealed that increases in perceived control post- tornado uniquely predicted decreases in weather-related conspiracy beliefs (see Figure 4).

204

Figure 4. Path analysis from change in perceived control to change in conspiracy beliefs

Table 3. Means and standard deviations and paired samples t-test results

M(SD) at Time M(SD) at Marginal t (df) p

1 Time 2 means (type

conspiracy)

Specific conspiracy 3.57 (1.89) 3.71(1.75) 3.64 (0.14) -1.73 (151) 0.09 theories

Generic conspiracy 3.64(1.48) 3.76(1.46) 3.70 (0.11) -1.32 (151) 0.19 theories

Weather related 2.06 (1.56) 1.96 (1.43) 2.01 (0.11) 1.12 (151) 0.26 conspiracies

205

Marginal means 3.09 (0.12) 3.14(0.11) / / / (wave)

2.81 (0.56) 2.91 (0.59) / -3.27 (151) 0.00 Perceived control

Affectedness 1.35 (1.87) 1.20 (1.75) 1.71 (151) 0.09

Discussion and Conclusion

Conspiracy theories continue to be strongly endorsed by the public (Jensen, 2013), with increasingly dangerous social consequences, including threats to health, the environment, and political engagement (Douglas, Sutton, Jolley & Wood, 2015). And while social media can legitimately be blamed for the ease of their propagation (Shao et al., 2018; del Vicario et al., 2016), scientific scrutiny of the underlying motivational causes is needed to understand their etiology and individuals’ receptivity to them. To that end, feelings of powerlessness and lack of control are often cited as motivating factors (e.g. Douglas, Sutton

& Cichocka, 2017), but the experimental evidence is mixed.

The current studies were therefore conducted to complement the existing experimental research on the effects of lack of control on conspiracy theory beliefs by examining this relationship in naturalistic settings. We conducted Study 1 against the real- world background of a political referendum on Macedonia’s name, a polarizing and personal issue that touched on a complex array of pragmatic concerns, intergroup identities, and interstate politics. Each of these factors is apparent in many conspiracy theories (Bale, 2007; van Prooijen & Douglas, 2018), but they are rarely, if ever, captured in the laboratory. In

206

Study 2, we examined participants’ perceptions of control and belief in conspiracy theories in the wake of a natural disaster.

The results of Study 1 indicated that people who opposed the name change, who presumably felt lower personal control, increased their conspiracy beliefs from 2018 to 2019, while the beliefs of those who supported the name change remained stable. Study 2 conceptually replicated these results, and in addition showed that effects were limited to conspiracy beliefs relevant to the control-threatening event.

It is not clear why the specificity of the effects differed between the two studies. One possibility is that the ambiguity and complexity of the threat in Study 1 was amenable to multiple interpretations, such that even “generic” conspiracies appeared relevant. For example, the statement that “Events throughout history are carefully planned and orchestrated by individuals for their own betterment,” intended as a generic statement about vague and unnamed forces, might have been given specific meaning in the context of the national referendum, which involved so many “events” and “individuals” that anyone could find aspects that appeared relevant to their interpretation of events. Tornadoes, in contrast, might be more constrained in their interpretation, permitting only limited extrapolation.

Alternatively, the effects of the tornadoes might have been more specific simply because more specific conspiracies were available for participants to use. Perhaps, that is, people restore control via whatever means are available, and while they gravitate toward theories with direct implications for the threatening events, more general claims will suffice when necessary.

This research is not without limitations, of course. Most obviously, an unavoidable aspect of the designs was that participants could not be randomly assigned to groups, leaving open the possibility that some other aspect besides low perceived control, could account for observed differences over time. Relatedly, there was no neutral control group, so it is not

207

certain whether low perceived control increased conspiracy beliefs, or whether high perceived control buffered against increases in them (or some combination of the two). It is also possible that beliefs would have changed even if the threatening events had not occurred, although there is no reason to suppose that this true, and, at least in Study 2, the role of tornado affectedness argues against it. In any event, we believe that these sacrifices to internal validity are more than earned back in terms of ecological validity. The current studies represent a rare look at conspiratorial thinking following the kind of major social upheaval or traumatic events that are thought to prompt it.

208

References:

Afifi, W. A., Afifi, T. D. & Merrill, A. (2014). Uncertainty and Control in the Context of a

Category‐Five Tornado. Research in Nursing and Health, 37, 358-366.

oi:10.1002/nur.21613

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.

Brehm, J.W. (1989). Psychological reactance: Theory and applications. Advances in

Consumer Research, 16, 72-75.

Brotherton, R., & French, C.C. (2015). Intention seekers: Conspiracist ideation and biased

attributions of intentionality. PLoS ONE 10(5): e0124125.

https://doi.org/10.1371/journal.pone.0124125

Brotherton, R., French, C., & Pickering, A. (2013). Measuring belief in conspiracy theories:

The generic conspiracist beliefs scale. Frontiers in Psychology, 4, 279.

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.

Cichocka, A., Marchlewska, M., & de Zavala, A. G. (2016). Does self-love or self-hate

predict conspiracy beliefs? Narcissism, self-esteem, and the endorsement of

conspiracy theories. Social Psychological and Personality Science, 7(2), 157–166.

209

del Vicario, M., Bessi, A., Zollo, F, Petroni, F., Scala, A., Caldarelli, G. Stanley, H.E.,

Quattrociocchi, W. (2016). Echo chambers in the age of misinformation. Proceedings

of the National Academy of Sciences, 113, 554-559.

Douglas, K. & Sutton, R. M. (2011). Does it take one to know one? Endorsement

of conspiracy theories is influenced by personal willingness to conspire. British

Journal of Social Psychology, 50, 544-552.

Douglas, K. M., Sutton, R. M., & Cichocka, A. (2017). The psychology of conspiracy

theories. Current Directions in Psychological Science, 26(6), 538–542.

Douglas, K., Sutton, R., Jolley, D., & Wood, M. (2015). The social, political, environmental,

and health-related consequences of conspiracy theories: problems and potential

solutions. In M. Bilewicz, A. Cichocka, and W. Soral (Eds.), The psychology of

conspiracy (pp.183-196). East Sussex: Routledge.

Goode, C., Keefer, L. A., & Molina, L. E. (2014). A compensatory control account of

meritocracy. Journal of Social and Political Psychology, 2, 313-334.

https://doi.org/10.5964/jspp.v2i1.372

Hart, J. & Graether, M. (2018). Something’s going on here. Psychological predictors of belief

in conspiracy theories. Journal of Individual Differences, 39 (4), 229-237.

Hauser, D.J., Ellsworth, P.C., & Gonzales, R. (2018). Are manipulation checks necessary?

Frontiers in Psychology, 9, 998. doi:10.3389/fpsyg.2018.00998

Imhoff, R. (2015). Beyond (right-wing) authoritarianism: Conspiracy mentality as an

incremental predictor of prejudice. In M. Bilewicz, A. Cichocka and W. Soral

(Eds.), The psychology of conspiracy (pp.122-141). East Sussex: Routledge.

210

Imhoff, R., & Lamberty, P. (2018). How paranoid are conspiracy believers? Toward a more

fine‐grained understanding of the connect and disconnect between paranoia and belief

in conspiracy theories. European Journal of Social Psychology. Advance online

publication.

Imhoff, R., & Lamberty, P. K. (2017). Too special to be duped: Need for uniqueness

motivates conspiracy beliefs. European Journal of Social Psychology, 47, 724-734.

https://doi.org/10.1002/ejsp.2265

Jensen, T. (2013). Democrats and Republicans differ on conspiracy theory beliefs. Retrieved

from http://www.publicpolicypolling.com/polls/democrats-and-republicans-differ-on-

conspiracy-theory-beliefs/

Kay, A. C., Sullivan, D., & Landau, M. J. (2015). Psychological importance of beliefs in

control and order: Historical and contemporary perspectives in social and personality

psychology. In M. Mikulincer, P. R. Shaver, E. Borgida, & J. A. Bargh (Eds.), APA

handbooks in psychology. APA handbook of personality and social psychology, Vol.

1. Attitudes and social cognition (pp. 309-337). Washington, DC, US: American

Psychological Association. http://dx.doi.org/10.1037/14341-010

Kay, A. C., Gaucher, D., Napier, J. L., Callan, M. J., & Laurin, K. (2008). God and the

government: Testing a compensatory control mechanism for the support of external

systems. Journal of Personality and Social Psychology, 95,18–35

Lantian, A., Muller, D., Nurra, C., & Douglas, K. M. (2017). “I know things they don’t

know!” the role of need for uniqueness in belief in conspiracy theories. Social

Psychology, 48, 160-173.

211

Laurin, K., Kay, A. C., & Moscovitch, D. A. (2008). On the belief in God : Towards an

understanding of the emotional substrates of compensatory control. Journal of

Experimental Social Psychology, 44, 1559–1562.

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, 109–117.

Nyhan, B., & Zeitzoff, T. (2018). Conspiracy and Misperception Belief in the Middle East

and North Africa. The Journal of Politics, 80(4), 1400-1404.

Orne, M.T. (2009). Demand Characteristics and the Concept of Quasi-Controls. In R.

Rosenthal and R.L. Rosnow (Eds.) Artifacts in behavioural research: Robert

Rosenthal and Ralph L. Rosnow’s Classic Books (pp. 110-138). Oxford: Oxford

University Press. Retrieved from:

http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780195385540.001.000

1/acprof-9780195385540-chapter-28

Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of Health and Social

Behavior, 19(1), 2-21. http://dx.doi.org/10.2307/2136319

Rosseel, Y. (2012). lavaan: An R Package for Structural Equation Modeling. Journal of

Statistical Software, 48(2), 1–36. http://www.jstatsoft.org/v48/i02/

Segal, K., Jong, J., & Halberstadt, J. (2018). The fusing power of natural disasters: An

experimental study. Self & Identity, 17, 574-586.

Shao, C., Hui, P.M., Wang, L., Jiang, X., Flammini A, Menczer, F., & Ciampaglia, G.L.

(2018). Anatomy of an online misinformation network. PLOS ONE 13(4),

e0196087.https://doi.org/10.1371/journal.pone.0196087.

212

Shepherd, S., Kay, A. C., Landau, M. J., & Keefer, L. A. (2011). Evidence for the specificity

of control motivations in worldview defense: Distinguishing compensatory control

from uncertainty management and terror management processes. Journal of

Experimental Social Psychology, 47, 949–958.

Smith, H. (2019, Jan 20). Thousands of Greeks protest over name change for Macedonia. The

Guardian. Retrieved from

https://www.theguardian.com/world/2019/jan/20/thousands-of-greeks-protest-over-

name-change-for-macedonia

Steenvoorden, E. H., & van der Meer, T. W. (2017). Continent of pessimism or continent of

realism? A multilevel study into the impact of macro-economic outcomes and

political institutions on societal pessimism, European Union 2006–2012. International

Journal of Comparative Sociology, 58(3), 192–214.

Stojanov, A., Bering, J.M., & Halberstadt, J. (under review). Does lack of control lead to

conspiracy theory beliefs?

Stojanov, A. & Halberstadt, J. (2019). The conspiracy mentality scale: Distinguishing

between irrational and rational suspicion. Social Psychology, 50(4), 215-232.

Stojanov, A. & Halberstadt, J. (under review). Does lack of control lead to conspiracy

beliefs? A meta-analysis.

Sullivan, D., Landau, M. J., & Rothschild, Z. K. (2010). An existential function of

enemyship : Evidence that people attribute influence to personal and political enemies

to compensate for threats to control. Journal of Personality and Social Psychology,

93, 434–449. https://doi.org/10.1037/a0017457

213

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. van Elk, M. & Lodder, P. (2018). Experimental manipulations of personal control do not

increase illusory pattern perception. Collabra, 4, 1–22. van Prooijen, J. W. (2016). Sometimes inclusion breeds suspicion: Self-uncertainty and

belongingness predict belief in conspiracy theories. European Journal of Social

Psychology, 46, 267- 279. https://doi.org/10.1002/ejsp.2157 van Prooijen, J.‐W. (2017). Why education predicts decreased belief in conspiracy theories.

Applied Cognitive Psychology, 31(1), 50-58. van Prooijen, J-W., & Acker, M. (2015). The influence of control on belief in conspiracy

theories: Conceptual and applied extensions. Applied Cognitive Psychology, 29, 753-

761. van Prooijen, J-W., & Douglas, K.M. (2017). Conspiracy theories as part of history: The role

of societal crisis situations. Memory Studies, 10(3), 323-333. van Prooijen, J-W., & Douglas, K.M. (2018). Belief in conspiracy theories: Basic principles

of an emerging research domain. European Journal of Social Psychology, 48 (7), 897-

908. van Prooijen, J.‐W., Douglas, K. M., & De Inocencio, C. (2018). Connecting the dots:

Illusory pattern perception predicts belief in conspiracies and the

supernatural. European Journal of Social Psychology, 48, 320–335.

214

Whitson, J. A., & Galinsky, A. D. (2008). Lacking control increases illusory pattern

perception. Science, 322, 115-117. https://doi.org/10.1126/science.1159845

215

Appendix A: Uncontrollable and controllable situations

Situation that participants experience as uncontrollable frequency Other people 85 Transport problems 48 Climate/weather 47 Prices 30 Sickness 21 Emotion 19 Loses 19 Politics 18 Queue /overcrowding 13 Lotto /chance(uncertainty) 12 Entertainment availability 11 Outages, (e.g. internet, power) 8 Body/bodily functions 8 Aging 7 Car accident 6 Genes 6 Space objects 5 Work schedule 5 Laws of physics 4 Love 3 Dreams 2 Miscellaneous 70 Invalid answers 23

216

Situations that participants experience as controllable frequency Consumption 64 Leisure time 61 Clothes 29 Sleeping 28 Work 27 Money 24 Attitude 21 Exercise 20 Shopping 18 Body 17 Emotions 16 Vacation 12 Friends 11 Objects 9 Where I live 7 Study 7 Who I date 6 Who I talk to on the phone 6 Religious activities 6 Children 6 Home Chores 5 Showering 5 Life 5 Who I vote for 4 Dog 4 Mode of transport 3 Gifts 3 Smoking 3

217

Habits 3 Favourite things 2 Media consumption 2 Miscellaneous 65

218

Chapter 6: General Discussion

Although conspiracy theories have been present throughout history, psychologists have only recently expanded their interest in them, partly as a result of the growing awareness of the potential harm they can cause (Jolley, 2013). Researchers have focussed not only on the consequences of beliefs in such theories (Douglas, Sutton, Jolley & Wood, 2015), but also in the psychological mechanisms that produce them, including personality traits, intergroup dynamics, and various individual motivations. Among the latter, the need for control has figured prominently in both research and theory, as well as in popular science communication

(e.g., Shermer, 2014; Oaklander, 2015, Yong, 2008).

Despite the attention to control as an explanatory variable, the logic of its causal effect on conspiracy beliefs is dubious, and the empirical evidence inconclusive. As noted in

Chapter 1, there are a number of reasons why conspiracy beliefs are unlikely candidates for restoring a threatened sense control, including their tendency for over-explanation and the social stigma that can accompany them. Moreover, “conspiracy beliefs” themselves have been operationalized in various ways, undoubtedly contributing to the variability in the observed effect of control on them.

Thus, the primary goal of this thesis was to systematically test the hypothesis that threats to perceived control lead to conspiracy beliefs. As a first step towards answering this question, in Chapter 2, I developed and validated a scale for measuring conspiracy mentality, abstracted from any specific conspiracy claims (which vary in their plausibility and generalizability). I then used that scale, sometimes in conjunction with other measures, in

Chapters 3 through 5, to test, using experimental, meta-analytic, and longitudinal methods, respectively, whether lack of control increases conspiracy beliefs.

219

The results from the experimental studies presented in Chapter 3 converged on a null finding: there was no evidence that lack of control, when manipulated, leads to conspiracy theory beliefs, even though measured control was negatively associated with the construct, as in previous research. This surprising result prompted a meta-analysis of all of the available experimental evidence, which largely supported the null findings, albeit with the caveat that the effect may be evident on specific rather than generic measures of conspiracy beliefs.

Finally, Chapter 5 examined the effect of lack of control on conspiracy beliefs in two ecologically valid settings – political upheaval and natural disaster –which additional tests showed are likely to be perceived as threatening to personal control. In the first study, conspiracy beliefs increased among those who opposed the change to Macedonia’s name, but not among those who supported it. In the second, those who experienced greater direct impact from a series of tornadoes were more likely to experience threats to control, which in turn predicted beliefs, but only about conspiracies directly related to the source of the threat (i.e., the weather). Furthermore, change in feelings of control over time predicted change in conspiracy theory beliefs, again only for threat-specific conspiracies. Taken together, the results presented in Chapter 5 suggest that, null effects in the laboratory notwithstanding, lack of control can increase conspiracy beliefs, but only under certain conditions, such as when strong personal threats can be linked with specific explanatory theories, though there are other relevant differences across the reported studies that may be important. For example, laboratory studies (including the current studies) measure conspiracy beliefs after a very short delay (typically within a standard one-hour experimental session), whereas the field studies in this thesis measured beliefs after weeks or months. Some studies suggest that threats to control may be only visible after a delay (Wichman, Brunner, & Weary, 2008), perhaps because due to an initial reaction is to inhibit control threatening thoughts. An initial defensive response would be consistent with similar effects in other threatening domains,

220

such as negative feedback (Martin, Tesser & McIntoch, 1993), unpleasant cognitions

(Wenzlaff & Wegner, 2000), and death anxiety (Greenberg, Pyszczynski, Solomon, Simon,

& Breus, 1994).

Differences in the time course of responses to threat may also explain another interesting finding in the current studies: the Macedonian name change produced changes in abstract conspiracy beliefs, but tornadoes’ effects were limited to very specific, threat- relevant conspiracy theories. One difference between the studies was that, in the former case, participants were surveyed after a six month delay, while in the latter, they were surveyed relatively soon (i.e. few days) after the threatening event. An intriguing possibility is that, in the immediate wake of a threat, participants are particularly drawn to specific conspiracy theories, which ultimately draw them into a larger world of conspiracy theory and culture. A sudden, severe illness, for example, might lead to speculation about the nefarious intentions of the pharmaceutical industry in the short term, which in turn acts as a gateway into like- minded others, who have “evidence” for more diverse claims along the same lines.

Limitations

Although I have focussed on discrepancies and variability in the construct and operationalization of conspiracy theories, the construct of “control” also has multiple dimensions and interpretations (Chipperfield, Perry & Stewart, 2012), and it is possible that different “domains” of control were examined in the experimental and naturalistic studies in this thesis. Bilewicz and Sedek (2015), for example, argue that the standard recall task manipulates control over one’s life overall, a domain about which participants tend to be confident (Lachman & Weaver, 1998). They may be less confident about the control they have over specific domains, such as their safety or security, which the tornadoes likely

221

challenged. More generally, different threats may produce different effects depending on the domain they challenge, and participants’ individual resilience in those domains.

In addition, the experimental and naturalistic “manipulations” may not have led to the same experience of threat. For example, there are some indications that when people initially lose control, they respond energetically to restore it, but after an extended period of control deprivation, they experience helplessness (Greenaway, Philipp & Storrs, 2017). Since recalling an uncontrollable experience is intended to create an immediate, acute challenge to participants’ sense of control, it may have been accompanied by a correspondingly immediate attempt to regain control, possibly limiting the effects of the manipulation. After days or months of chronically threatened control, participants may have been more vulnerable to conspiracy explanations. Thus, it may be because of, rather than despite of, a six month delay that oppositional Macedonians found themselves entertaining thoughts of conspiracies.

It is also worth noting that most participants in the reported studies were online participants, mostly Mechanical Turk workers. While there are some concerns raised in the literature regarding use of MTurk samples, such as familiarity with similar paradigms

(Landers & Behrend, 2015), low compensation rates (Landers & Behrend, 2015), and the lack of oversight over what participants are doing simultaneously (Fleischer, Mead & Huang,

2015), there are also advantages, such as greater diversity and intrinsic motivation relative to student samples (Buhrmester, Kwang & Gosling, 2011; Paolacci & Chandler, 2014).

Furthermore, the concerns with online samples can largely be managed, as I did in the current studies by including attention check questions (Dance, 2015), recruiting only participants with high approval rates, and permitting participation in only one study in a given line of research. Nevertheless, it is possible that some participants had undergone the experimental manipulations before (e.g., if they had been involved in similar research in the past),which is known to attenuate the effect size (Chandler, Paolacci, Peer, Mueller, & Ratliff, 2015).

222

Perhaps, then, using other or more “naïve” samples would have produced stronger or different results in my experimental paradigms.

As with all research, the current studies represented a trade off between experimental control and external validity, and naturalistic studies have their own vulnerabilities. Anxiety, for example, is an obvious potential confound when studying natural disasters (Afifi, Afifi, &

Merrill, 2014), and, arguably, political change. Although I controlled for arousal and positivity in the “name change” study, those measures may not have been adequate proxies for the anxiety participants felt.

Implications

The findings may have implication for theory. If the relationship between lack of control and conspiracy beliefs is not domain-general, but domain-specific, as the naturalistic studies seem collectively to suggest, it would imply that the lack of control in a particular domain threatens perceptions of order in that domain, and compensatory efforts to restore perceptions of control and order will be focussed, and more effective, in that same domain.

Thus, following a random, devastating weather event, believing that the pharmaceutical industry deliberately produces harmful vaccines may not be as appealing a source of compensatory order as believing that the weather is manipulated by the government. Future studies could systematically test these ideas about domain specificity, for example by depriving participants in different domains of control (e.g. health vs politics) and measuring belief in health and political conspiracies, or endorsement of health vs political institutions.

Depending on the results, future instantiations of compensatory control theory may include the notion of domain specificity of control threats and compensatory attempts.

223

Directions for future research

The current studies were primarily concerned with common cognitive mechanisms that might explain conspiracy beliefs, mostly ignoring individual differences. However, there are a number of personality variables that are relevant to both the control and conspiracy ideation constructs, which could be fruitful avenues for further research. For example, individuals differ in their chronic need for (and chronic feeling of) control, and examination of conspiracy beliefs among these individuals could provide insights into both the beliefs and the individuals, as well as represent an important moderator of experimental manipulations of control. Relatedly, some personality types (e.g., schizotypy) and clinical conditions (e.g., paranoid schizophrenia) are characterized by need for control or strategies to establish order

(e.g., the tendency to see unusual patterns in the world), and individuals with these conditions may be more (or less) susceptible to external control threats.

In addition to differences among individuals, there are also differences among control threats. For example, the same “uncontrollable” event may have different effects if judged as uncontrollable by anyone (e.g., a natural disaster), versus uncontrollable only by the self.

Further, the plausibility of a given theory, the availability of alternative explanations, and the availability of other means of restoring control, are all likely to contribute to a given theory’s success. For example, believing that 9/11 was caused by the government, when another plausible explanation is available (i.e., that 9/11 was caused by Al Qaeda) seems as likely to threaten than to buttress (some) people’s sense of order and structure. On the other hand, conspiracy theories that pertain to events that appear more or less random, such as natural disasters, might be more attractive, especially if other external systems of control are unavailable.

There are also theoretical reasons to think that the link between control and conspiracy beliefs is bi-directional. From the perspective of compensatory control

224

theory (Kay, Whitson, Gaucher & Galinsky, 2009), although belief in conspiracy theories satiates a higher order need for order and structure, it should also reduce the need to perceive personal control. Additionally, to the extent that conspiracy theories imply that people are merely a puppet on a string pulled by the conspirators, people who endorse conspiracy theories should experience cognitive dissonance (Festinger, 1957): they cannot see themselves both as marionets and as agentic selves, potentially leading to a reduced sense of personal control. Some empirical research supports these contentions. For example, participants exposed to pro-conspiracy material, compared to those exposed to anti- conspiracy material, scored higher on a measure of political powerlessness (Jolley &

Douglas, 2014a; Jolley et al., 2014b). Similarly, Douglas and Leite (2018) found that participants were more disengaged (i.e. showed higher turnover intentions) after they were experimentally exposed to organizational conspiracy theories.

The possibility that conspiracy beliefs threaten personal control suggests other, more sinister downstream effects, as well as new lines of inquiry for researchers. Not all efforts to restore control are benign, particularly if other, socially sanctioned activities, such as demonstrations or activism, are condemned in conspiracy theories as part of a master plan to keep the population docile and under control (Franks, Bangerter, Bauer, Hall & Noort, 2017).

Individuals endorsing such theories may therefore view extremism as the only way to restore agency. Of course, only a small minority of even hardcore conspiracy theorists are prone to commit violence, and research should focus on the moderators that are likely to trigger such behaviour, such as anomie (Abalakina-Paap et al., 1999). However, even among the peaceful majority, widespread propagation of control-threatening conspiracy theories would likely have negative effects on trust in institutions and lead to endorsing systems that promise to restore order, such as extremist ideologies.

225

If conspiracy beliefs indeed satiate a higher order need for structure and order, and function in a hydraulic fashion with perceived control and other external control systems, this suggests at last two mechanisms for reducing conspiracy beliefs. First, enhancing individuals’ perceived control could reduce individuals’ need for conspiratorial accounts (van Prooijen,

2019). So far there is only one (underpowered) study comparing baseline to heightened perceptions of control (van Prooijen & Acker, 2015), but the results were promising: the participants in the high control condition demonstrated lower conspiracy beliefs compared to those in a neutral condition. However, in my own studies, as well as those of others (e.g.

Cutright, 2012) it has proven difficult to increase perceptions of control from a baseline level, so it is not yet clear whether enhancing control is a feasible strategy of combatting conspiracy beliefs. Even if perceptions of high control are inherently difficult to evoke, restoring participants chronically deprived levels of perceived control to baseline might be sufficient.

The changes over time in the “tornado” study, for example, could be seen as the result of the natural restoration of control, previously damaged by the traumatic effects of the storms.

Strengthening the perception that existing institutions function effectively could be another avenue for reducing conspiracy beliefs. There are indeed some indications that social structure and conspiracy ideation may be inversely related. For example, in open societies that practice transparency and that are characterized by freedom of the press, an independent legal system, a strong civil society, and clear expectations about how society functions, conspiracy theories are less likely to flourish (Bruder, Haffke, Neave, Nouripanah, & Imhoff;

Moore, 2016), compared to societies where people feel powerless, experience passivity, have an impaired sense of personal responsibility, and where feelings of uncertainty prevail

(Bruder et al., 2013; Stojanov & Douglas, in preparation), possibly because in the latter societies conspiracy theories provide some predictability, order and a convenient framework for the interpretation of current events.

226

Finally, perceptions of control and order are only some of many human motivations that may drive conspiracy beliefs. Other potential factors that future research may explore are self-esteem maintenance or group identification. For example, once people start believing in a conspiracy theory, do they continue to believe in it for self-esteem maintenance purposes and belongingness to a group? Given the many different, sometimes conflicting needs people have (Talevich, Read, Walsh, Iyer, & Chopra, 2017), it is also important to explore under what conditions some of these needs take primacy over others and lead to conspiracy beliefs

(e.g., when the need for uniqueness overwhelms the need to see the world accurately).

New questions that expand the research horizons can also be asked. For example, questions such as “why people stop believing in conspiracy theories” or “what are the effects of writing fake news on the writer?” or “under what conditions people come up with conspiracy explanations spontaneously” could also provide insights about so far neglected aspect of conspiracy thinking. Future research can also dig deeper into the layers of conspiracy thinking by looking at “second order conspiracy theories” (conspiracy theory about a conspiracy theory e.g. “The government deliberately encourages conspiracy theories about 9/11 so that it can distract from the real problems”) and what factors – conspiracy theory ideation or scepticism will predict such beliefs.

Conclusion

Although this thesis was intended to address the question, “Does lack of control lead to conspiracy theories”, the results suggest that the more appropriate question is, “Under what conditions does lack of control leads to conspiracy beliefs”. Tentatively, the answers include a minimum intensity for the control-threatening event, a specific link between the threat and the conspiratorial explanation of it and, perhaps, a delay between the threatening event and the opportunity to restore control. This account suggests a more limited effect of

227

control than some theoretical accounts suggest, but a limited effect is not an inconsequential one. Indeed, any inclination to endorse conspiracy theories, even just to restore local and specific threats to control, might open the door towards a world of other conspiracy beliefs, both specific and generic, ultimately leading to higher propensity for conspiracy thinking.

228

Referenecs:

Abalakina-Paap, M., Stephan, W., Craig, T., & Gregory, W. (1999). Beliefs in

Conspiracies. Political Psychology, 20(3), 637-647. Retrieved from

www.jstor.org/stable/3792165

Afifi, W. A., Afifi, T. D. & Merrill, A. (2014). Uncertainty and Control in the Context of

a Category‐Five Tornado. Research in Nursing and Health, 37, 358-366.

doi:10.1002/nur.21613

Bilewicz, M., & Sedek, G. (2015). Conspiracy stereotypes: Their sociopsychological

antecedents and consequences. In M. Bilewicz, A. Cichocka, & W. Soral (Eds.), The

psychology of conspiracy (p. 3–22). London: Routledge/Taylor & Francis Group.

Bruder, M., Haffke, P., Neave, N., Nouripanah, N., and Imhoff, R. (2013). Measuring

individual differences in generic beliefs in conspiracy theories across cultures:

conspiracy mentality questionnaire. Front. Psychol. 4:225. doi:

10.3389/fpsyg.2013.00225

Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s Mechanical Turk: A New

Source of Inexpensive, Yet High-Quality, Data? Perspectives on Psychological

Science, 6(1), 3–5. https://doi.org/10.1177/174569161039398

Chandler, J., Paolacci, G., Peer, E., Mueller, P., & Ratliff, K. A. (2015). Using Nonnaive

Participants Can Reduce Effect Sizes. Psychological Science, 26(7), 1131–

1139. https://doi.org/10.1177/0956797615585115

Chipperfield, J. G., Perry, R. P., & Stewart,. T. L. (2012). Perceived Control. In. V. S.

Ramachandran's (Ed.) Encyclopedia of Human Behaviour (p.42-48). Boston, MA:

Elsevier

229

Cutright, K. M. (2012). The Beauty of Boundaries: When and Why We Seek Structure in

Consumption. Journal of Consumer Research, 38, 775-790.

https://doi.org/10.1086/661563

Dance, A. (2015). How online studies are transforming psych research. Proceedings of the

National Academy of Sciences , 112 (47), 14399-

14401; DOI: 10.1073/pnas.1520769112

Douglas, K. M., Leite, A. C. (2017). Suspicion in the workplace: Organizational conspiracy

theories and work-related outcomes. British Journal of Psychology, 108, 486–506.

Douglas, K. M., Sutton, R. M., Jolley, D., & Wood, J.M. (2015). The social, political,

environmental, and health consequences of conspiracy theories. In M. Bilewicz, A.

Ciichocka, and W. Soral (Eds), The Psychology of Conspiracy (pp. 183-200).

London, UK: Routledge.

Festinger, L. (1957). A theory of cognitive dissonance. Stanford, California : Stanford

University Press.

Fiske, S. & Neuberg, S.L. (1990). A Continuum of Impression Formation, from Category-

Based to Individuating Processes: Influences of Information and Motivation on

Attention and Interpretation. Advances in Experimental Social Psychology, 23, 1-74.

Fleischer, A., Mead, A. D., & Huang, J. (2015). Inattentive responding in MTurk and other

online samples. Industrial and Organizational Psychology, 8(2), 196-202.

doi:10.1017/iop.2015.25

230

Franks, B., Bangerter, A., Bauer, M.W., Hall, M. and Noort, M.C. (2017) Beyond

“Monologicality”? Exploring Conspiracist Worldviews. Front. Psychol. 8:861. doi:

10.3389/fpsyg.2017.00861

Giry, J. (2017). A specific social function of rumors and conspiracy theories: strengthening

community’s ties in troubled times. A multilevel analysis. Slovensky

Narodopis, 65(2), 187-202. Retrieved from

https://search.proquest.com/docview/1917879287?accountid=14700

Greenaway, K. H., Philipp, M. C., & Storrs, K. (2017). The motivation for control: Loss of

control promotes energy, effort, and action. In M. Bukowski, M. Kofta, & I. Fritsche.

(Eds.), Coping with lack of control in a social world (p. 35–48). New

York: Routledge.

Greenberg, J., Pyszczynski, T., Solomon, S., Simon, L., & Breus, M. (1994). Role of

consciousness and accessibility of death-related thoughts in mortality salience effects.

Journal of Personality and Social Psychology, 67, 627-637.

Grzesiak-Feldman, M. (2013). The effect of high-anxiety situations on conspiracy

thinking. Current Psychology: A Journal for Diverse Perspectives on Diverse

Psychological Issues, 32(1), 100–118.

Jolley, D. (2013). Are conspiracy theories just harmless fun? Psychologist, 26(1), 60-62.

Jolley, D., & Douglas, K.M. (2014a). The effects of anti-vaccine conspiracy theories on

vaccination intentions. PLOS ONE, 9(2): e89177. doi:

10.1371/journal.pone.0089177;

Jolley, D., & Douglas, K.M. (2014b). The social consequences of conspiracism: Exposure to

conspiracy theories decreases the intention to engage in politics and to reduce one’s

231

carbon footprint. British Journal of Psychology, 105, 35-56. doi:

10.1111/bjop.12018

Jones, E. E., & Davis, K. E. (1965). From acts to dispositions the attribution process in

person perception. Advances in Experimental Social Psychology, 2, 219-266.

Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system justification theory:

Accumulated evidence of conscious and unconscious bolstering of the status

quo. Political Psychology, 25(6), 881-919. doi:10.1111/j.1467-9221.2004.00402.x

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. https://doi.org/10.1111/j.1467-

8721.2009.01649.x

Kreko, P. (2015). Conspiracy theory as collective motivated cognition. In M. Bilewicz, A.

Cichocka, & W. Soral, (Eds.) The Psychology of Conspiracy (p. 62-75). London:

Routledge.

Kruglanski, A. & Webster, D. (1996). Motivated Closing of the Mind: "Seizing" and

"Freezing”. Psychological Review, 103 (2), 263-283.

Kruglanski, A. (1989). Lay Epistemics and Human Knowledge : Cognitive and Motivational

Bases. Boston, MA : Springer.

Landers, R. N., & Behrend, T. S. (2015). An inconvenient truth: Arbitrary distinctions

between organizational, mechanical turk, and other convenience samples. Industrial

and Organizational Psychology, 8(2), 142-164. doi:10.1017/iop.2015.13

Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology,

32(2), 311-328.

232

Lerner. M. (1980). The belief in a just world: a fundamental delusion. New York: Plenum

Press.

Martin, L.L, Tesser, A., & McIntosh, W.D (1993). Wanting but not having: The effects of

unattained goals on thoughts and feelings. In D.M. Wegner

and J.W. Pennebaker (Eds.), The handbook of mental control (p. 552-572). New

York, NY: Prentice Hall.

Moore, A. (2016). Conspiracy and Conspiracy Theories in Democratic Politics. Critical

Review, 28(1), 1- 23. doi: 10.1080/08913811.2016.1178894

Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a

Participant Pool. Current Directions in Psychological Science, 23(3), 184–

188. https://doi.org/10.1177/0963721414531598

Roberts, J.M. (1972). The mythology of the secret societies. London: Sacker & Warburg.

Seligman, M. (1975). Helplessness: on depression, development, and death. San Francisco:

W. H. Freeman.

Skinner, E.A. (1995). Perceived control, motivation and coping. Thousand Oaks, CA: Sage.

Stojanov, A. & Douglas, K. (in preparation). Conspiracy beliefs in Britain and Macedonia.

Talevich, J. R., Read, S. J., Walsh, D. A., Iyer, R., & Chopra, G. (2017). Toward a

comprehensive taxonomy of human motives. PloS One, 12(2), e0172279.

doi:10.1371/journal.pone.0172279

Van Prooijen, J.-W. (2019). Empowerment as a tool to reduce belief in conspiracy theories.

In J. Uscinski (Ed.), Conspiracy theories and the people who believe in them (pp. 432-

442). New York, NY: Oxford University Press.

233

van Prooijen, J-W, & Douglas, K. (2017). Conspiracy theories as part of history: The role of

societal crisis situations. Memory Studies, 10(3), 323-333.

Wenzlaff, R.M., & Wegner, D.M. (2000). Thought suppression. Annual Review of

Psychology, 51, 59-91.

Wichman, A. L., Brunner, R. and Weary, G. 2008. Immediate and delayed effects of causal

uncertainty inductions on uncertainty accessibility. Journal of Experimental Social

Psychology, 44: 1106–1113.

234