Exploring the Mechanisms that Allow Incompatible Beliefs to Coexist in the Cognitive System

Abstract

Despite assumptions that people strive for consistency between their beliefs, endorsement of mutually incompatible ones is not rare - a tendency we have previously labelled doublethink, by borrowing from Orwell. In an attempt to understand the nature of doublethink and the underlying mechanism that enables incompatible beliefs to coexist, we conducted two preregistered studies (total N = 691). To do so, in Study 1, we first explored how doublethink relates to (1) thinking styles (rational/intuitive, actively open-minded thinking, and need for cognitive closure), (2) a set of irrational beliefs (magical health, conspiratorial, superstitious, and paranormal beliefs) and (3) its predictiveness for questionable health practices (non-adherence to medical recommendations and use of traditional, complementary and alternative medicine). We then additionally expanded the set of health behaviors in Study 2, and related doublethink to trust in two epistemic authorities - science and the wisdom of the common man. Finally, in both studies, we explored whether those prone to inconsistent beliefs are also more likely to simultaneously rely on conventional and alternative medicine, despite their apparent incompatibility. While doublethink was positively related to need for cognitive closure and different irrational beliefs that easily incorporate contradictions, as well as negatively to actively open-minded thinking, we did not find it to be predictive of the use of non-evidence-based medicine nor of its simultaneous use with official medicine. It seems that this novel construct can be best understood as a feature of the cognitive system that allows incompatible claims to enter it. However, once beliefs are within the system, they are compartmentalized, without any cross-referencing between them. This is further reflected in non-evidence-based beliefs persisting within the belief system, irrespective of their content.

Keywords

Doublethink inconsistent beliefs irrational beliefs health behaviors trust in science thinking styles

Introduction

Despite seminal psychological works (Festinger, 1957; Heider, 1946) assuming consistency is a default state within the belief system, and inconsistency an unpleasant state urging to be resolved, there is mounting empirical evidence this might not always be the case, i.e. that a person can concurrently endorse mutually incompatible beliefs. There are, however, individual differences in this proneness to doublethink (Petrović & Žeželj, 2022, 2023), more often found in people with a more intuitive and less rational approach to information processing. Doublethink is content-independent, observed in different beliefs, regardless of their type. In other words, it is a general characteristic of the belief system. There are clusters of concrete beliefs, however, in which inconsistency is more likely to be found than in others. Within the literature, those are labeled as irrational, epistemically suspect, or unwarranted beliefs (Pennycook et al., 2015; Rizeq et al., 2021), and inconsistency is sometimes even listed as one of their key features (Lewandowsky & Cook, 2020). While there is evidence that ties doublethink to this set of beliefs (Lazarević et al., 2023; Petrović & Žeželj, 2022, 2023), it is still under-researched. Understanding this link might be particularly important as irrational beliefs can have detrimental consequences, most notably leading to the use of questionable health practices (e.g. Knežević et al., 2024; Lazarević et al., 2023; Lindeman et al., 2023; Oliver & Wood, 2014; Purić et al., 2022, 2023).

Across two studies, we explored how doublethink relates to different types of irrational beliefs, how reflective it is of a certain type of superficial information processing, and whether it can be predictive of risky health behaviors.

Irrational Beliefs and Superficial Information Processing

Irrational beliefs include a wide range of beliefs that in the broadest sense do not comply with postulates of normative rationality, and contradict scientific knowledge or ontological assumptions about reality (Žeželj & Lazarević, 2019). These beliefs are more and more examined together rather than individually. Apart from their stated base conceptual similarity, all different types of irrational beliefs also tend to empirically cluster together and are highly interrelated, forming three main domains - conspiratorial, pseudoscientific and paranormal (e.g. Lobato et al., 2014; Rizeq et al., 2021; Šrol, 2022; Teovanović et al., 2024). Within such a belief system, contradictory beliefs might be particularly likely to arise. This is especially true for conspiratorial beliefs, where people are likely to endorse mutually incompatible conspiratorial narratives (Lukić et al., 2019; Petrović & Žeželj, 2022, 2023; Wood et al., 2012). Recent findings further corroborate doublethink’s stable positive correlation with conspiratorial beliefs (both general and specific), but also demonstrate its relation to other types of irrational beliefs, including superstition and magical health beliefs (Knežević et al., 2024; Lazarević et al., 2023). This might be because the irrational belief system is organized in such a way that it tolerates the existence of inconsistencies (Wood, 2017).

This tolerance of contradictions might be a reflection of a more superficial information processing style (i.e. relying more on intuitive, but less on rational thinking) that is typical for people who endorse irrational beliefs (e.g., Knežević et al., 2024; Lazarević et al., 2023; Pennycook et al., 2015; Svedholm & Lindeman, 2013; Yelbuz et al., 2022). Similarly, another characteristic of such a belief system might be a strong intolerance of uncertainty reflected in the need for cognitive closure - it has previously been tied to conspiratorial thinking (Marchlewska et al., 2018), pseudoscientific claims (Van Elk, 2019), and paranormal beliefs (Wilt et al., 2022). The relation of doublethink to the need for cognitive closure is, however, less straightforward and has not been examined in the literature previously. Given that doublethink can be considered a “high entropy” (more chaotic) thinking style, it should imply less need for cognitive closure - such a disheveled belief system suggests more tolerance of uncertainty.

While doublethink cannot be labeled as one of the irrational beliefs, given its lack of content and that it hinders on the relation between beliefs specifically, it does seem to fit well into the irrational mindset, allowing for very different, and even incompatible beliefs to coexist. However, which general style of information processing shapes doublethink and irrational beliefs and if it’s one and the same is still left to be explored.

Detrimental Health Consequences of Doublethink and irrational Beliefs

Endorsement of irrational beliefs might be particularly problematic if we take into account its consequences, especially in the domain of health. To maintain their health or to cure themselves, people resort to various non-evidence-based practices or turn away from evidence-based ones. Previous studies (Purić et al., 2022, 2023) exploring the relation of the irrational mindset to questionable health practices have systematized these behaviors into two broad groups - (1) intentional non-adherence to medical recommendations (iNAR; e.g. avoiding checkups, self-medication, etc.) and (2) use of traditional, complementary and alternative medicine (TCAM; e.g. homeopathy, use of herbal remedies, etc.). And while the irrational mindset was an important predictor of TCAM, but not iNAR (Knežević et al., 2024; Lazarević et al., 2023; Purić et al., 2022, 2023), the relation of doublethink to these practices was less clear. One study found that doublethink was positively, albeit weakly, related to use of TCAM only (Lazarević et al., 2023), while another found no relation of doublethink to either of the two types of questionable health practices (Knežević et al., 2024). Given its relation to the irrational mindset, it might be another candidate for predicting the use of questionable health practices, and its relation to questionable health practices thus requires more scrutiny.

Study 1

Aims and Hypotheses

In this first of two studies, we aimed firstly to replicate and check the robustness of previously found relations of doublethink to various irrational beliefs (Knežević et al., 2024; Lazarević et al., 2023; Petrović & Žeželj, 2022, 2023). To this end, we focused on the relations of doublethink to conspiratorial, paranormal, superstitious, and magical health beliefs, thus capturing all three main domains of irrational beliefs (conspiratorial, pseudoscientific and paranormal; Rizeq et al., 2021; Šrol, 2022). We expected that doublethink would be positively related to conspiratorial (H1a), paranormal (H1b), superstitious (H1c), and magical health beliefs (H1d), and we also expected that all irrational beliefs would be positively related to each other (H2), implying the existence of an irrational mindset.

To go even further, doublethink has been previously related to less analytical and more intuitive thinking style and actively open-minded thinking (Petrović & Žeželj, 2022, 2023) and the same is true for irrational beliefs (e.g., Pennycook et al., 2015; Svedholm & Lindeman, 2013; Yelbuz et al., 2022). We aimed to then explore how both doublethink and irrational beliefs are tied to a specific, more superficial approach to information processing, by examining their relations to actively open-minded thinking, rational and intuitive thinking styles, and the need for cognitive closure. To replicate previous findings on doublethink’s relation to superficial information processing, we expected that doublethink would be positively related to intuitive thinking (H3a), but negatively to actively open-minded (H3b) and rational thinking (H3c). In this study we also focused on another specific characteristic of thinking style that has not been tied to doublethink previously - need for cognitive closure. We expected it to also be negatively related to doublethink (H3d). Similarly, as found in previous studies, we expected irrational beliefs to be positively related to intuitive thinking (H4a), but negatively to actively-open minded (H4b) and rational thinking (H4c).

In the domain of health behaviors, we expected doublethink to be related to less adherence to official medical recommendations (H5a) and more use of alternative medicine practices (H5b). We also anticipated that irrational beliefs and doublethink would contribute positively to the prediction of non-adherence to official medical recommendations (H6a) and use of alternative medicine (H6b), over and above measures of cognitive style. Finally, we expected that (H7) doublethink would be positively related to the simultaneous endorsement of official and alternative medicine.

Figure 1 shows an overview of the study.

Figure 1.

Three main domains explored in the context of doublethink within the study.

We preregistered all the hypotheses, analyses, and the sampling plan here - https://aspredicted.org/qe4h2.pdf.

Methods

Materials

Doublethink was assessed using the Proneness to doublethink scale (Petrović & Žeželj, 2022). The scale assesses the tendency to accept mutually contradictory beliefs and consists of 11 pairs of such beliefs (e.g. Some people are essentially irreparable and Every single person is capable of growth and change). The participants assess their agreement with each item individually on a scale from 1 (completely disagree) to 4 (completely agree). The scale is presented in two separate blocks of 11 items each (one item from each pair), alongside two buffer items in each block, and with other scales as buffers between the two blocks. The score is calculated by adding up the number of contradictory pairs where a person agrees with (marked 3 or 4 on the scale) both items in a pair. The scale shows good construct and convergent validity (e.g. Lazarević et al., 2023; Petrović & Žeželj, 2022, 2023), and is available in open access along with the scoring procedure - https://osf.io/udth2/.

To assess the conspiratorial subdomain of epistemically suspect beliefs, we used the Conspiracy Mentality Questionnaire (Bruder et al., 2013), which assesses the general propensity to engage in conspiratorial thinking. We wanted to focus on general tendency towards conspiratorial ideation, instead of focusing on any one specific conspiracy theory to avoid any content overlap with our other measures. The scale consists of five items (e.g. Government agencies closely monitor all citizens.) to which participants responded using a 5-point scale (1 - Completely disagree; 5 - Completely agree). The score is calculated as the mean of answers across all items. The scale is psychometrically sound and validated cross-culturally (Imhoff et al., 2022); the Serbian version previously showed good factorial validity and reliability (Milošević-Đorđević et al., 2021).

To assess paranormal beliefs, we used two measures. First, paranormal beliefs were assessed via the Revised Paranormal Beliefs scale (Tobacyk, 2004), with 26 items (one reverse coded) separated into seven subscales - traditional religious beliefs, psi beliefs, witchcraft, superstition, spiritualism, extraordinary life forms and precognition (e.g. A person’s thoughts can influence the movement of a physical object.) The participants indicated their agreement with the items on a 7-point scale (1 - Completely disagree; 7 - Completely agree). In line with previous findings (e.g., Houran et al., 2001; Lange et al., 2000) showing that a two-factor structure of the scale is more adequate, we calculated the scores for these two subscales - New Age Philosophy (11 items) and Traditional Paranormal Beliefs (5 items). According to Lange and colleagues (2000) the first subscale captures new age paranormal abilities such as astrology, while the second focuses on traditional supernatural concepts such as witchcraft. Item breakdown per subscale is available on OSF - https://osf.io/udth2/.

As an additional measure of the paranormal subdomain of irrational beliefs, we included superstition, to make sure to tap into culturally relevant paranormal beliefs as well. Superstition was assessed via the Superstition scale (Žeželj et al., 2009), consisting of 20 items that measure superstitious beliefs and behaviors (e.g. I am pleased when my palm itches, it might mean money is on the way.). Seven items in the scale are reversely coded. The participants responded using a 5-point scale (1 - Completely disagree, 5 - Completely agree), and we calculated the score as a mean of answers on all items, given that the scale has one main factor (Žeželj et al., 2009).

To assess the domain of pseudoscientific beliefs, we included magical health beliefs which were measured with the 10 items from the general factor of the Magical Beliefs about Food and Health Scale (Lindeman et al., 2000). We opted for this scale because it captures the general principles of pseudoscientific beliefs well, without explicitly mentioning pseudoscientific practices. The participants indicated their agreement with the items (e.g. An imbalance between energy currents lies behind many illnesses.) on a 5-point scale (1 - completely disagree; 5 - completely agree). The score was calculated as a mean of all items. Both the original Finnish (e.g. Aarnio & Lindeman, 2004), as well as the English (e.g. Bryden et al., 2018) and the Serbian (e.g. Lazarević et al., 2023) versions of the scale have good reliability (alpha over .85).

Actively open-minded thinking was assessed with the eight-item version of the Actively Open-Minded Thinking Scale (Bronstein et al., 2019; Stanovich & West, 1997). We opted for this version of the scale, given that some other short operationalizations show low reliability (see Petrović & Žeželj, 2023 for details). The scale measures the tendency to use evidence to form and revise beliefs (Stanovich & West, 1997), and contains five reversely coded items. The participants answered using a 6-point scale (1- Completely disagree, 6 - Completely agree). We calculated the score by taking the mean on all items.

Need for cognitive closure was assessed with the short version of the Need for Cognitive Closure scale (Roets & Van Hiel, 2011; Webster & Kruglanski, 1994). The scale measures the desire for certainty and consists of 15 items (e.g. When I have made a decision, I feel relieved). The participants indicated their agreement with the items on a 6-point scale (1- Completely disagree, 6 - Completely agree), and we calculated the score as a mean of answers on all items. The short version of the scale was empirically validated and psychometrically sound (Roets & Van Hiel, 2011).

Health behaviors. To measure how often people resort to questionable health practices, we assessed both intentional non-adherence to official medical recommendations and the use of traditional, complementary, and alternative medicine.

To assess intentional non-adherence to medical recommendations, we took four of the most frequent non-adherence behaviors reported in a previous study (e.g. It had happened to me that I took antibiotics even though a doctor did not prescribe them to me.; Purić et al., 2023). The participants indicated on a binary scale (0 - no, 1 - yes) if the outlined behavior has happened to them in the last 12 months. Given that the elimination of the item regarding the use of sunscreen would lead to a significant change in reliability (from α = .49 to α = .57), we omitted this item and were left with three items. We calculated the score as a proportion of the behaviors that have happened to the participant.

To assess TCAM use, we took two items from each of the four TCAM domains (Purić et al., 2022) - alternative medical systems (e.g. homeopathy), mind-body practices (e.g. meditation), natural product-based practices (e.g. vitamins, minerals, and antioxidants), and rituals/customs (e.g. praying for health). For each practice, the participants indicated if they had used it in the last 12 months, and we calculated the score as a proportion of used practices.

Finally, to assess proneness to the simultaneous use of TCAM and conventional medicine, we included the following item - Which type of medicine do you prefer to use when you need to solve a health issue? The participants answered on a 7-point scale (−3 - official medicine, zero - both equally, 3 - Traditional, complementary and alternative medicine). We recoded the answers to absolute values and then inversely coded them, so that higher numbers indicate a higher proneness to simultaneous use. The score then ranged from 1 to 4.

Sampling

Our final sample consisted of a total of N = 306 participants from Serbia, which was in line with our preregistered a priori power analysis (see the preregistration for details - https://aspredicted.org/qe4h2.pdf). A total of 476 participants from Serbia took part in the study, which was run online. They were recruited using snowballing and Facebook and Twitter ads. In line with the preregistration, we removed 170 participants who failed one of the five attention checks. Per preregistration, we checked for speeders via the relative speed index (which indicates how fast participants finish the survey compared to the median participant; RSI, Leiner, 2019). However, there were no participants with an RSI >2. The sample consisted of 90.5% women, 8.5% men, while the rest of the participants chose the other/rather would not say option. The mean age of the participants was 48.02 (SD = 13.02).

Results

All data and analytical code is available on the Open Science Framework: https://osf.io/udth2/. Data was analyzed using R version 4.3 (R Core Team, 2024).

Table 1 details the means, standard deviations, and correlations of doublethink, irrational beliefs, and measures of information processing style.

Table 1.

Means, standard deviations, reliability, and correlations of doublethink, irrational beliefs, and information processing style (Study 1).

Variable	M	SD	1	2	3	4	5	6	7	8	9	10
1. Doublethink (0–11)	4.28	2.37	0.73
2. Conspiracy mentality (1–5)	3.95	0.75	.31***	0.77
3. Magical health beliefs (1–5)	2.86	0.89	.35***	.59***	.89
4. New age philosophy paranormal beliefs (1–7)	2.96	1.39	.13	.42***	.62***	.91
5. Traditional paranormal beliefs (1–7)	2.99	1.51	.18*	.41***	.54***	.75***	.82
6. Superstition (1–5)	2.08	0.72	.24***	.28***	.40***	.55***	.55***	.87
7. Actively open-minded thinking (1–6)	4.78	0.87	−.33***	−.21**	−.26***	−.26***	−.28***	−.38***	.79
8. Need for cognitive closure (1-6	4.11	0.78	.30***	.17*	.10	.02	.10	.27***	−.30***	.84
9. Rational thinking style (1–5)	4.05	0.82	−.06	−.06	−.06	−.09	−.18*	−.22**	.15	−.18*	.75
10. Experiential thinking style (1–5)	3.39	0.97	.08	.25***	.31***	.34***	.28***	.22**	−.09	−.19*	−.02	0.77

Note. M and SD are used to represent mean and standard deviation, respectively. The main diagonal shows reliability of all measures (GLB for doublethink (Petrović & Žeželj, 2022) and Cronbach’s alpha for all other measures).

All p values were adjusted for multiple observations using Holm’s method.

* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

All of our hypotheses regarding irrational beliefs and doublethink were confirmed. As expected (H1a-d), doublethink was positively related to all irrational beliefs (except for one paranormal beliefs subscale), while all irrational beliefs were positively related to each other, in line with our prediction (H2).

Moreover, as we expected, doublethink was negatively related to actively open-minded thinking (H3b). However, contrary to our predictions (H3d), the relation of doublethink to the need for cognitive closure was positive. Doublethink was also, unexpectedly, not related to either the intuitive (H3a) or rational (H3c) thinking style.

As for irrational beliefs, they were all positively related to intuitive thinking, in line with our predictions (H4a), while as expected, negatively related to actively open-minded thinking (H4b). Rational thinking style was negatively related to only traditional paranormal and superstitious beliefs, while other correlations were not significant, thus only partially in line with H4c.

To explore how doublethink is related to health behaviors, we also examined their interrelations (Table 2).

Table 2.

Means, Standard Deviations, and Correlations of doublethink and Different health Behaviors - Study 1.

Variable	M	SD	1	2	3
1. Doublethink (0–11)	4.28	2.37	.73
2. Intentional non-adherence to medical recommendations (0–1)	0.55	0.34	.11	.57
3. Use of TCAM (0–1)	0.29	0.16	−.01	.14	.39
4. Preference for simultaneous use of conventional and alternative medicine (1–4)	2.16	1.35	.04	.17*	.36***

Note. For 4., higher scores indicate more preference for simultaneous use of conventional and alternative medicine. M and SD are used to represent mean and standard deviation, respectively. The main diagonal shows the reliability of all measures (GLB for doublethink (Petrović & Žeželj, 2022) and Cronbach’s alpha for all other measures).

All p values were adjusted for multiple observations using Holm’s method.

* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

As Table 2 suggests, we found no relation of doublethink to either intentional non-adherence (H5a) or use of TCAM (H5b), which was not in line with our expectations. Similarly, we did not find a significant correlation between doublethink and preference for the simultaneous use of official and alternative medicine (H7).

It should be noted, however, that the short measures of intentional non-adherence and TCAM use had low reliability (α = .57 and α = .39, respectively). This is why we also explored how doublethink is related to individual non-adherence behaviors, as well as individual TCAM practices. After correcting for multiple comparisons using Holm’s method, none of the correlations of doublethink to individual behaviors were significant.

To account for the poor reliability of some of the measures, we also report correlations corrected for attenuation (see Supplement S1 on OSF - https://osf.io/udth2/). As expected, this lead to an increase in the magnitude of all correlations, however, the non-significant correlations remain at a level that does not reach significance. The overall pattern of results thus remains the same.

Finally, to assess the relative contribution of doublethink and irrational beliefs to the prediction of intentional non-adherence and use of TCAM when information processing style is taken into account, we ran a path analysis. We opted for this for the sake of clarity and to take into account multiple comparisons. We fit the model using the maximum likelihood estimation with robust standard errors and a Satorra-Bentler scaled test statistic. We included intentional non-adherence and TCAM as the outcomes, and all measures of superficial information processing and irrational beliefs, as well as doublethink as predictors (Table 3).

Table 3.

Path Analysis With TCAM and iNAR as Outcomes - Study 1.

Outcome	Predictor	Estimate	Standardized estimate	p
iNAR
	Actively open minded thinking	−0.017	−.042	.526
	Need for cognitive closure	0.010	.023	.721
	Rational thinking style	0.019	.045	.443
	Experiential thinking style	0.007	.020	.753
	Doublethink	0.010	.068	.308
	Conspiracy mentality	0.011	.024	.738
	Magical health beliefs	−0.021	−.054	.518
	New age philosophy beliefs	0.005	.018	.856
	Traditional paranormal beliefs	0.040*	.178*	.043
	Superstition	−0.005	−.010	.899
TCAM
	Actively open minded thinking	0.015	.078	.153
	Need for cognitive closure	0.025	.119	.060
	Rational thinking style	−0.005	−.025	.614
	Experiential thinking style	0.003	.015	.781
	Doublethink	−0.011**	−.164**	.006
	Conspiracy mentality	−0.024	−.113	.049
	Magical health beliefs	0.063***	.349***	<.001
	New age philosophy beliefs	0.009	.075	.319
	Traditional paranormal beliefs	0.018*	.167*	.024
	Superstition	0.016	.073	.244

Note. Significant predictors are printed in bold.

* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

As Table 3 suggests and as opposed to our predictions (H5a), almost none of the predictors contributed to the prediction of intentional non-adherence, with the exception of traditional paranormal beliefs, which only marginally contributed to the prediction (standardized estimate = .178, p = .043). The model explained around 5% of the variance of intentional non-adherence (R² = .047).

For TCAM use (Table 3), the results were partially in line with our predictions (H5b). While none of the thinking style variables contributed significantly to prediction, irrational beliefs and doublethink did. Magical health beliefs (standardized estimate = .349, p < .001) and traditional paranormal beliefs (standardized estimate = .167, p = .024) contributed positively to the prediction, as expected. Doublethink, however, contributed negatively to the prediction (standardized estimate = −.164, p = .006). In total, the model explained around a total of 24% of the variance of TCAM practices’ use.

Given that we preregistered a test of our hypotheses via hierarchical linear regression and that those results were fully in line with the model presented here, for the sake of brevity, we report those results in Supplement S1.

Robustness Checks

Given that some of the distributions were skewed, we also reran the analyses with normalized scores. The pattern of the results remained the same (see Supplementary file S2 - https://osf.io/udth2/ for more information).

Study 2

Introduction

While our preliminary findings suggest that there is no relation between doublethink and questionable health behaviors, it is also possible that doublethink could be tied to the use of health practices in less direct ways. Cognitive polyphasia (Jovchelovitch & Gervais, 1999) is a concept closely related to doublethink referring to a state where contradicting explanations of the world are endorsed by the same individual. Much of the research on cognitive polyphasia has been in the health domain, where studies show that, for example, practices and beliefs from traditional medicine are combined with conventional medical knowledge, despite them often conflicting with each other (Jovchelovitch & Priego-Hernández, 2015). Thus, endorsement of contradictory beliefs might not be predictive of proneness to questionable health behaviors as such, but could potentially explain how people use different questionable health practices. For example, people can use TCAM alternatively to official medicine, simultaneously with it, or even preventively. Doublethink could then be related specifically to simultaneous use and can explain how people rely on different practices simultaneously, even when they are based on contradictory principles.

Moreover, doublethink could stoke trust in unwarranted epistemic authorities, or hinder trust in appropriate ones, much like irrational beliefs do. This could then further endorse the use of questionable health practices. One such warranted epistemic authority is science. Mistrust in science has been repeatedly tied to conspiratorial (Rutjens & Većkalov, 2022), but also to pseudoscientific (Fasce & Picó, 2019a) and paranormal beliefs (Fasce & Picó, 2019b). It has not, however, been explored alongside doublethink so far. Furthermore, during the pandemic, mistrust in science has been repeatedly tied to questionable health practices, both in the domain of non-adherence to conventional practices and in the domain of TCAM use (Algan et al., 2021; Pagliaro et al., 2021, Žeželj et al., 2023).

Conversely, often contrasted with trust in science (Merkley & Loewen, 2021) is the so-called trust in the “wisdom of the common/ordinary man” (Žeželj et al., 2023), i.e. trust in the common sense of an average layperson. Trust in the wisdom of the common man has also been related to epistemically suspect beliefs, in particular belief in conspiracy theories, and to questionable health practices (Žeželj et al., 2023). Still, its relation to doublethink has not been examined previously. Not only that, but despite them being contrasted in literature (e.g. this referral of trust to the ordinary people as in Oliver & Rahn, 2016 has been equated with anti-intellectualism), and their implicit incompatibility, trust in science and trust in the “wisdom of the common man” were previously shown to be unrelated in non-US samples (Žeželj et al., 2023). This means that, at least for some people, these two types of epistemic authorities are not mutually exclusive, and doublethink could again help reconcile them.

Aims and Hypotheses

Given that the reliability of health behaviors was low in the previous study, in this study we focused on expanding the list of health behaviors and used the full original instruments (Purić et al., 2022, 2023) to discern better if doublethink is related to questionable health practices. We again expected that doublethink would be positively related to less adherence to official medical recommendations (H1a) and the use of alternative medicine practices (H1b). Moreover, drawing from previous findings that non-adherence to medical recommendations and use of TCAM are positively related (Purić et al., 2022, 2023), we also explored if doublethink moderates this relation - for those higher on doublethink relying on both conventional and alternative medicine might be easier. We thus expected that doublethink would moderate the relation between non-adherence to medical recommendations and use of TCAM, so that this correlation is weaker for people more prone to doublethink (H2). We also expected that doublethink would be positively related to the perception of conventional and alternative medicine as complementary (H3a), and to simultaneous reliance on conventional and alternative medicine when having to solve a health problem (H3b).

We also focused on exploring how doublethink is related to trust in two types of epistemic authorities - science and the wisdom of the common man. We expected that doublethink would be negatively related to trust in science (H4a), but positively to trust in the “wisdom of the common man” (H4b), in line with previous findings on the relation of these two types of trust to irrational beliefs (Žeželj et al., 2023). Moreover, we expected that doublethink would moderate the relation between these two types of trust, so that this relation is stronger and positive for higher scores of doublethink (H5).

We again preregistered all the hypotheses, analyses, and the sampling plan here: https://aspredicted.org/2ms8j.pdf.

Methods

Materials

Intentional non-adherence was assessed with the iNAR-12 questionnaire (Purić et al., 2023). The questionnaire consists of 12 items that assess lifetime intentional non-adherence to different medical recommendations (e.g. taking antibiotics without prescription, not reporting all symptoms to a doctor), to which participants answer using a binary scale (0 - it has never happened to me; 1 - it has happened to me). The score is calculated by averaging all items.

Use of TCAM practices was assessed with the TCAM-22 questionnaire (Purić et al., 2022), which consists of 22 alternative medical practices (e.g. homeopathy, use of herbal tinctures, prayer, use of crystals). Responses are given in a binary format (0 - I have never used this practice, 1 - I have used this practice or I am still using it), and a total score is calculated by averaging all items. Both of the scales (iNAR and TCAM) were empirically derived from a larger pool of items, and are psychometrically sound and reliable (e.g. Knežević et al., 2024; Lazarević et al., 2023).

To more precisely assess the perception of the relationship between alternative and conventional medicine we used both an attitudinal indicator and a behavioral one.

To measure attitudes towards this relationship (attitudinal indicator), we constructed five items for this study (e.g. For me, alternative and conventional medicine complement each other). The participants indicated their agreement with the items on a 7-point scale (1 - completely disagree; 7 - completely agree). All items had high loadings on a single factor (>.80), and the scale had an excellent reliability (α = .94), so we calculated the score by averaging answers on all five items.

To assess if participants are inclined to also use the two types of medicine simultaneously (behavioral indicator), we again asked them what type of medicine they prefer to use when they have to resolve a certain health issue. The participants responded using a 7-point scale (−3 - conventional medicine; zero - both equally; +3 - alternative medicine). We recoded the answers to absolute values and then inversely coded them, so that the highest score (i.e. 3) reflected that participants were prone to simultaneously use, while lower scores indicated that they are more ready to opt for one of the two types. The score ranged from zero to 3.

To assess trust in epistemic authorities we used two items (Žeželj et al., 2023), in which we asked the participants how much trust they have in (a) science and (b) the wisdom of the “common man”, when it comes to knowledge about and dealing with health problems. The participants answered using a 7-point scale (1 - no trust at all; 7 - trust completely).

Doublethink was assessed as in the previous study.

Sampling

Our final sample size, which was in line with our a priori power analysis (see preregistration for details - https://aspredicted.org/2ms8j.pdf), consisted of 385 participants from Serbia, collected online using snowballing and Facebook ads. As per preregistration, three attention checks were used as filters, so those that failed them were filtered out and could not continue on with the study (127 participants failed the first attention check, three the second one, and three the final one). We again checked for speeders via the relative speed index (RSI, Leiner, 2019) and removed a total of five participants that had an RSI above 2. The sample consisted of 80.5% women, 17.6% men, while the rest of the participants chose the other/rather would not say option. The mean age of the participants was 49.85 (SD = 11.91).

Results

All data and analytical code are available on the Open Science Framework: https://osf.io/udth2/. Data was again analyzed using R version 4.3 (R Core Team, 2024).

Table 4 details means, standard deviations, and correlations, as well as reliabilities for all multi-item measures on the main diagonal.

Table 4.

Means, standard deviations, and correlations of all used variables - Study 2.

Variable	M	SD	1	2	3	4	5	6
1. Doublethink (0–11)	4.10	2.25	.68
2. iNAR (0–1)	0.36	0.21	.06	.68
3. TCAM (0–1)	0.37	0.17	.01	.23***	.79
4. Trust in science (1–7)	5.84	1.19	.01	−.17**	−.20***	/
5. Trust in the wisdom of the common man (1–7)	4.34	1.51	.20***	.11	.34***	−.06	/
6. Relationship between alternative and conventional medicine - attitudinal indicator (1–7)	4.65	1.79	.24***	.09	.42***	−.10	.38***	.94
7. Relationship between alternative and conventional medicine - behavioral indicator (0–3)	1.08	1.21	.10	.14	.35***	−.22***	.23***	.51***

Note. For 6. and 7., higher scores indicate that conventional and alternative medicine are seen/used as more compatible. M and SD are used to represent mean and standard deviation, respectively. The main diagonal shows reliability of all measures (GLB for doublethink (Petrović & Žeželj, 2022) and Cronbach’s alpha for all other measures).

All p values were adjusted for multiple observations using Holm’s method.

* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

While both the iNAR and TCAM use measures showed improved reliability relative to Study 1, contrary to our predictions, doublethink was not related to either iNAR (H1a) or TCAM use (H1b). Those more prone to doublethink were more likely to perceive conventional and alternative medicine as complementary (H3a), but, unexpectedly, not more prone to actually use them simultaneously when they had a health issue (H3b). We found no correlation of doublethink and trust in science (H4a), but as expected, those higher on doublethink were more likely to have trust in the wisdom of the “common man” when it came to resolving health problems (H4b). We also found that trust in science was related to more adherence to recommendations, and less use of TCAM, while trust in the wisdom of the “common man” positively contributed only to the prediction of use of TCAM.

As in Study 1, due to the lower reliability of some measures, we also report correlations corrected for attenuation in Supplement S1 - https://osf.io/udth2/. As in Study 1, this increases the magnitude of the correlations, but did not change the pattern of results reported.

Moderation Analyses

To explore if doublethink moderates the relation between iNAR and use of TCAM (H3), as well as the relation between two types of trust (H5), we ran two moderation analyses using a path model with maximum likelihood estimation with robust standard errors and a Satorra-Bentler scaled test statistic using the R package laavan (Rosseel, 2012). Contrary to our prediction, doublethink was not a significant moderator of either of those two relationships.

There was a significant main effect found between iNAR and TCAM (beta = .235, p < .001), however, there was no significant main effect of doublethink on TCAM use (beta = −.002, p = .969). There was also no moderating effect of doublethink (beta = −.008, p = .864).

Similarly, there was no moderating effect of doublethink on the relation between trust in science and the wisdom of the “common man” (beta = −.019, p = .743). We also did not find a main effect of trust in science (beta = −.063, p = .273), while there was a significant main effect of doublethink (beta = 0.204, p < .001).

Robustness Checks

As in Study 1, we reran the analyses with normalized scores and without outliers as a robustness check. The pattern of the results remained largely the same (see Supplementary file S3 - https://osf.io/udth2/ on OSF for more information).

Discussion

Across two studies, we explored how doublethink relates to superficial information processing and irrational beliefs, and whether it is predictive of questionable health practices and reflected in trust in different epistemic authorities.

Our results show that, while doublethink is not related to trust in science, it is related to higher trust in the “wisdom of the common man”. This type of epistemic authority is often referred to as anti-intellectualism and is implicitly, but also often explicitly contrasted with trust in experts or even used as a proxy for distrust in experts (e.g., Merkley, 2020; Merkley & Loewen, 2021; Oliver & Rahn, 2016). However, despite this tendency to equate them in the literature, we again found no correlation between the trust in science and in the wisdom of the common man in a Serbian sample (Žeželj et al., 2023). We also find that this trust in the common man is related to doublethink. This implies that, at least in our non-Western samples, trust in the ordinary man is not incompatible with trust in experts, or that incompatibility can be overcome within the belief system.

We have now also consistently observed the expected positive relations between doublethink and endorsement of irrational beliefs, very different in content at that. This is in line with our previous findings on the positive link between doublethink and conspiratorial beliefs (Petrović & Žeželj, 2022, 2023), and more recent findings on its positive relation to other types of irrational beliefs as well (Lazarević et al., 2023). Moreover, we again found that those prone to doublethink are also more likely to be more actively open-minded in their thinking, in line with previous findings (Petrović & Žeželj, 2023). The pattern of doublethink’s relation to the set of irrational beliefs and the set of thinking styles does not allow us to firmly situate it within either of those blocks. On one hand, doublethink implies a departure from principles of normative reasoning - one of the defining features of irrational beliefs. However, irrational beliefs are content-specific, and differ from each other specifically in that content, while doublethink is largely content-independent, and can arise in any type of belief as long as they are incompatible. On the other hand, doublethink can be considered a thinking style that implies a higher tolerance of contradictory beliefs (Petrović & Žeželj, 2022). In contrast to other thinking styles which are assessed through self-report and thus represent a measure of a preferred approach to information processing, doublethink is a direct measure of the number of contradictory beliefs a person adopts. In our view, this cannot be dismissed as only a method difference, but signals an important difference in the nature of these constructs.

Having in mind the nature of doublethink, we expected that such a belief system should be made possible by a higher tolerance of uncertainty, i.e. lower need for cognitive closure. The relation we observed was, nevertheless, positive. This could be due to the fact that this belief system is permeable only upon entry, when beliefs are indiscriminately adopted. Once adopted, however, beliefs may rarely be cross-referenced and subjected to scrutiny which makes them harder to correct or change altogether, which is typical for those high in need for cognitive closure (Webster & Kruglanski, 1994). It could also be that drawing from a need for certainty and cognitive closure, people adopt mutually exclusive beliefs to “cover their bases as wide as possible” and represent all possible outcomes within their belief system.

Finally, despite our expectations, we did not find that doublethink is predictive of the use of questionable health practices. Those prone to doublethink were, however, more likely to see conventional and alternative medicine as mutually compatible, but not to also simultaneously use them. This would suggest that while doublethink is related to beliefs on alternative and conventional medicine, this relation does not necessarily transfer to behaviors as well. Moreover, while out of the scope of this study, the pattern of results suggests the possibility of doublethink contributing to TCAM use indirectly, by stoking trust in inadequate epistemic authorities and especially irrational beliefs, given its stable relation to them.

Limitations and Future Directions

While we aimed to cover a wider range of irrational beliefs, we certainly cannot claim it is representative of the whole set. Future studies could, for example, focus on examining whether doublethink is related to pseudoscientific beliefs specifically, as they are one of the three key domains of irrational beliefs (Rizeq et al., 2021; Šrol, 2022).

Our studies used a correlational design, which did not allow us to make assumptions about the direction of relations between the constructs. One could try to experimentally manipulate doublethink and observe the outcomes of the potential change - for example, by highlighting the cognitive dissonance in people’s own beliefs, and then testing if this subsequently reduces irrational beliefs.

In these studies we tested the relationship between doublethink and basic thinking styles, while future ones could relate it to basic personality dispositions, especially focusing on Openness to experience.

Since it was out of the scope of this study, we did not check the participants’ overall health status. It could be that those with poorer overall health or those currently suffering from medical illnesses have more opportunities to use both conventional medicine and TCAM. Future studies could thus examine whether those groups in particular are more likely to simultaneously rely on both, and whether proneness to doublethink is higher for those who do.

Finally, in this paper, we examined trust in two epistemic authorities independently of each other and found that doublethink only relates to the trust in the wisdom of the common man. It could be informative to look into whether there is a trade-off between these two types of trust by directly contrasting them, similarly to how we contrasted alternative and conventional medicine within a single question. This would allow us to see if people choose science over the wisdom of the common man or vice versa, or whether they rely on both epistemic authorities simultaneously.

Conclusions

In two highly powered preregistered studies we observed robust relationships between a relatively novel concept of doublethink and superficial information processing style, as well as a set of content-varied irrational beliefs - all in line with previous research on the topic (Petrović & Žeželj, 2022, 2023). We developed a way to meaningfully measure individual differences in endorsement of incompatible beliefs, find a non-trivial portion of the population who score relatively high on it, and offer evidence for it to be considered a feature of a specific mindset. Having in mind the consistency paradigm that was pervasive in psychology in the previous period, this should be taken as a signal that doublethink should be further researched and understood better.

Footnotes

Acknowledgment

This work is a part of Marija Petrović’s PhD thesis entitled “Consistently inconsistent: Predictivity and validity of doublethink” at the University of Belgrade, supervised by Dr Iris Žeželj.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia [451-03-47/2023-01/200163].

Ethical Statement

ORCID iDs

Marija B. Petrović

Iris Žeželj

Data availability statement

The data that support the findings of this study are openly available on the Open Science Framework at:

Author Biographies

Marija Petrović is a PhD candidate and a research assistant at the Department of Psychology at the University of Belgrade. Her main research interest are irrational beliefs, with a special focus on endorsement of contradictory beliefs (i.e. doublethink), which is also the topic of her dissertation.

Iris Žeželj is a full professor of social psychology at the Department of Psychology at the University of Belgrade. She studies different forms of irrational beliefs and their behavioral consequences, and has supervised the first author's thesis.

References

Aarnio

Lindeman

(2004). Magical food and health beliefs: A portrait of believers and functions of the beliefs. Appetite, 43(1), 65–74. https://doi.org/10.1016/j.appet.2004.03.002

Algan

Cohen

Davoine

Foucault

Stantcheva

(2021). Trust in scientists in times of pandemic: Panel evidence from 12 countries. Proceedings of the National Academy of Sciences, 118(40), e2108576118. https://doi.org/10.1073/pnas.2108576118

Bronstein

M. V.

Pennycook

Bear

Rand

D. G.

Cannon

T. D.

(2019). Belief in fake news is associated with delusionality, dogmatism, religious fundamentalism, and reduced analytic thinking. Journal of Applied Research in Memory and Cognition, 8(1), 108–117. https://doi.org/10.1037/h0101832

Bruder

Haffke

Neave

Nouripanah

Imhoff

(2013). Measuring individual differences in generic beliefs in conspiracy theories across cultures: Conspiracy mentality questionnaire. Frontiers in Psychology, 4(1), 225. https://doi.org/10.3389/fpsyg.2013.00225

Bryden

G. M.

Browne

Rockloff

Unsworth

(2018). Anti-vaccination and pro-CAM attitudes both reflect magical beliefs about health. Vaccine, 36(9), 1227–1234. https://doi.org/10.1016/j.vaccine.2017.12.068

Fasce

Picó

(2019a). Conceptual foundations and validation of the pseudoscientific belief scale. Applied Cognitive Psychology, 33(4), 617–628. https://doi.org/10.1002/acp.3501

Fasce

Picó

(2019b). Science as a Vaccine: The relation between scientific literacy and unwarranted beliefs. Science & Education, 28(1–2), 109–125. https://doi.org/10.1007/s11191-018-00022-0

Festinger

(1957). A theory of cognitive dissonance. Stanford University Press.

Heider

(1946). Attitudes and cognitive Organization. Journal of Psychology, 21(1), 107–112. https://doi.org/10.1080/00223980.1946.9917275

10.

Houran

Irwin

H. J.

Lange

(2001). Clinical relevance of the two-factor rasch version of the revised paranormal belief scale. Personality and Individual Differences, 31(3), 371–382. https://doi.org/10.1016/S0191-8869(00)00143-4

11.

Imhoff

Zimmer

Klein

António

J. H. C.

Babinska

Bangerter

Bilewicz

Blanuša

Bovan

Bužarovska

Cichocka

Delouvée

Douglas

K. M.

Dyrendal

Etienne

Gjoneska

Graf

Gualda

Hirschberger

Van Prooijen

J.-W.

(2022). Conspiracy mentality and political orientation across 26 countries. Nature Human Behaviour, 6(3), 392–403. https://doi.org/10.1038/s41562-021-01258-7

12.

Jovchelovitch

Gervais

M.-C.

(1999). Social representations of health and illness: The case of the Chinese community in England. Journal of Community & Applied Social Psychology, 9(4), 247–260. https://doi.org/10.1002/(SICI)1099-1298(199907/08)9:4<247::AID-CASP500>3.0.CO;2-E

13.

Jovchelovitch

Priego-Hernández

(2015). Cognitive polyphasia, knowledge encounters and public spheres. In Sammut

Andreouli

Gaskell

Valsiner

(Eds.), The Cambridge Handbook of Social Representations (pp. 163–178). Cambridge University Press. https://doi.org/10.1017/CBO9781107323650.014

14.

Knežević

Petrović

Ninković

Zupan

Purić

Lukić

Živanović

Teovanović

Stanković

Žeželj

(2024). Lifetime prevalence of questionable health behaviors and their psychological roots: A preregistered nationally representative survey. PsyArXiv. [Preprint]. https://doi.org/10.31234/osf.io/zf4q7

15.

Lange

Irwin

H. J.

Houran

(2000). Top-down purification of Tobacyk’s revised paranormal belief scale. Personality and Individual Differences, 29(1), 131–156. https://doi.org/10.1016/S0191-8869(99)00183-X

16.

Lazarević

L. B.

Knežević

Purić

Teovanović

Petrović

M. B.

Ninković

Živanović

Stanković

Branković

Lukić

Opačić

Žeželj

(2023). Tracking variations in daily questionable health behaviors and their psychological roots: A preregistered experience sampling study. Scientific Reports, 13(1), 14058. https://doi.org/10.1038/s41598-023-41243-w

17.

Leiner

D. J.

(2019). Too fast, too Straight, too Weird: Non-Reactive indicators for Meaningless data in Internet surveys. Survey Research Methods, 13(3), 3. https://doi.org/10.18148/srm/2019.v13i3.7403

18.

Lewandowsky

Cook

(2020). The conspiracy theory Handbook. https://sks.to/conspiracy

19.

Lindeman

Keskivaara

Roschier

(2000). Assessment of magical beliefs about food and health. Journal of Health Psychology, 5(2), 195–209. https://doi.org/10.1177/135910530000500210

20.

Lindeman

Svedholm-Häkkinen

A. M.

Riekki

T. J. J.

(2023). Searching for the cognitive basis of anti-vaccination attitudes. Thinking & Reasoning, 29(1), 111–136. https://doi.org/10.1080/13546783.2022.2046158

21.

Lobato

Mendoza

Sims

Chin

(2014). Examining the relationship between conspiracy theories, paranormal beliefs, and pseudoscience acceptance Among a University population: Relationship between unwarranted beliefs. Applied Cognitive Psychology, 28(5), 617–625. https://doi.org/10.1002/acp.3042

22.

Lukić

Žeželj

Stanković

(2019). How (ir)rational is it to believe in contradictory conspiracy theories? Europe's Journal of Psychology, 15(1), 94–107. https://doi.org/10.5964/ejop.v15i1.1690

23.

Marchlewska

Cichocka

Kossowska

(2018). Addicted to answers: Need for cognitive closure and the endorsement of conspiracy beliefs. European Journal of Social Psychology, 48(2), 109–117. https://doi.org/10.1002/ejsp.2308

24.

Merkley

(2020). Anti-intellectualism, populism, and Motivated Resistance to expert Consensus. Public Opinion Quarterly, 84(1), 24–48. https://doi.org/10.1093/poq/nfz053

25.

Merkley

Loewen

P. J.

(2021). Anti-intellectualism and the mass public’s response to the COVID-19 pandemic. Nature Human Behaviour, 5(6), 706–715. https://doi.org/10.1038/s41562-021-01112-w

26.

Milošević-Đorđević

Žeželj

Đurić

Ž.

(2021). Beyond general political attitudes: Conspiracy mentality as a global belief system predicts endorsement of international and local conspiracy theories. Journal of Social and Political Psychology, 9(1), 144–158. https://doi.org/10.5964/jspp.5609

27.

Oliver

J. E.

Rahn

W. M.

(2016). Rise of the Trumpenvolk: Populism in the 2016 Election. The Annals of the American Academy of Political and Social Science, 667(1), 189–206. https://doi.org/10.1177/0002716216662639

28.

Oliver

J. E.

Wood

(2014). Medical conspiracy theories and health behaviors in the United States. JAMA Internal Medicine, 174(5), 817–818. https://doi.org/10.1001/jamainternmed.2014.190

29.

Pagliaro

Sacchi

Pacilli

M. G.

Brambilla

Lionetti

Bettache

Bianchi

Biella

Bonnot

Boza

Butera

Ceylan-Batur

Chong

Chopova

Crimston

C. R.

Álvarez

Cuadrado

Ellemers

Formanowicz

Zubieta

(2021). Trust predicts COVID-19 prescribed and discretionary behavioral intentions in 23 countries. PLoS One, 16(3), e0248334. https://doi.org/10.1371/journal.pone.0248334

30.

Pennycook

Allan Cheyne

Barr

Koehler

Fugelsang

(2015). On the reception and detection of pseudo-profound bullshit. Judgment and Decision Making, 10(6), 549–563. https://doi.org/10.1017/s1930297500006999

31.

Petrović

Marija

Žeželj

Iris

. (2022). Thinking inconsistently: Development and validation of an instrument for assessing proneness to doublethink. European Journal of Psychological Assessment. https://doi.org/10.1027/1015-5759/a000645

32.

Petrović

Marija

Žeželj

Iris

. (2023). Both a bioweapon and a hoax: The curious case of contradictory conspiracy theories about COVID-19. Thinking & Reasoning. https://doi.org/10.1080/13546783.2022.2088618

33.

Purić

Petrović

M. B.

Živanović

Lukić

Zupan

Branković

Ninković

Lazarević

L. B.

Stanković

Žeželj

(2023). Development of a novel instrument for assessing intentional non-adherence to official medical recommendations (iNAR-12): A sequential mixed-methods study in Serbia. BMJ Open, 13(6), e069978. https://doi.org/10.1136/bmjopen-2022-069978

34.

Purić

Živanović

Petrović

Lukic

Knezevic

Teovanovic

Ninković

Lazić

Opacic

Branković

Lazarevic

L. B.

Zezelj

(2022). Something old, something new, something borrowed, something green: How different domains of traditional, alternative, and complementary medicine use are rooted in an irrational mindset. PsyArXiv. [Preprint]. https://doi.org/10.31234/osf.io/agp5y

35.

R Core Team . (2024). R: A language and environment for statistical computing. Computer Software. https://www.r-project.org/

36.

Rizeq

Flora

D. B.

Toplak

M. E.

(2021). An examination of the underlying dimensional structure of three domains of contaminated mindware: Paranormal beliefs, conspiracy beliefs, and anti-science attitudes. Thinking & Reasoning, 27(2), 187–211. https://doi.org/10.1080/13546783.2020.1759688

37.

Roets

Van Hiel

(2011). Item selection and validation of a brief, 15-item version of the Need for Closure Scale. Personality and Individual Differences, 50(1), 90–94. https://doi.org/10.1016/j.paid.2010.09.004

38.

Rosseel

(2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

39.

Rutjens

B. T.

Većkalov

(2022). Conspiracy beliefs and science rejection. Current Opinion in Psychology, 46(2), 101392. https://doi.org/10.1016/j.copsyc.2022.101392

40.

Šrol

(2022). Individual differences in epistemically suspect beliefs: The role of analytic thinking and susceptibility to cognitive biases. Thinking & Reasoning, 28(1), 125–162. https://doi.org/10.1080/13546783.2021.1938220

41.

Stanovich

K. E.

West

R. F.

(1997). Reasoning independently of prior belief and individual differences in actively open-minded thinking. Journal of Educational Psychology, 89(2), 342–357. https://doi.org/10.1037/0022-0663.89.2.342

42.

Svedholm

A. M.

Lindeman

(2013). The separate roles of the reflective mind and involuntary inhibitory control in gatekeeping paranormal beliefs and the underlying intuitive confusions. British Journal of Psychology, 104(3), 303–319. https://doi.org/10.1111/j.2044-8295.2012.02118.x

43.

Teovanovic

Purić

Živanović

Lukic

Branković

Stanković

Knezevic

Lazarevic

L. B.

Zezelj

(2024). The role of cognitive biases in shaping irrational beliefs: A multi-study Investigation. PsyArXiv. [Preprint]. https://doi.org/10.31234/osf.io/g59k4

44.

Tobacyk

J. J.

(2004). A revised paranormal belief scale. International Journal of Transpersonal Studies, 23(1), 94–98. https://doi.org/10.24972/ijts.2004.23.1.94

45.

Van Elk

(2019). Socio-cognitive biases are associated to belief in neuromyths and cognitive enhancement: A pre-registered study. Personality and Individual Differences, 147(6), 28–32. https://doi.org/10.1016/j.paid.2019.04.014

46.

Webster

D. M.

Kruglanski

A. W.

(1994). Individual differences in need for cognitive closure. Journal of Personality and Social Psychology, 67(6), 1049–1062. https://doi.org/10.1037/0022-3514.67.6.1049

47.

Wilt

Stauner

May

R. W.

Fincham

F. D.

Pargament

K. I.

Exline

J. J.

(2022). Who engages with supernatural Entities? An Investigation of personality and cognitive style predictors. Imagination, Cognition and Personality, 41(4), 373–414. https://doi.org/10.1177/02762366211065677

48.

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. https://doi.org/10.1111/bjop.12231

49.

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. https://doi.org/10.1177/1948550611434786

50.

Yelbuz

B. E.

Madan

Alper

(2022). Reflective thinking predicts lower conspiracy beliefs: A meta-analysis. Judgment and Decision Making, 17(4), 720–744. https://doi.org/10.1017/S1930297500008913

51.

Žeželj

Lazarević

L. B.

(2019). Irrational beliefs. Europe's Journal of Psychology, 15(1), 1–7. https://doi.org/10.5964/ejop.v15i1.1903

52.

Žeželj

Pavlović

Vladisavljević

Radivojević

(2009). Construction and behavioral validation of superstition scale. Psihologija, 42(2), 141–158. https://doi.org/10.2298/PSI0902141Z

53.

Žeželj

Petrović

Ivanović

Kurčubić

(2023). I trust my immunity more than your vaccines: “Appeal to nature” bias strongly predicts questionable health behaviors in the COVID-19 pandemic. PLOS ONE, 18(2), e0279122. https://doi.org/10.1371/journal.pone.0279122