Do as the Romans do: On the authoritarian roots of pseudoscience

Recent research highlights the implications of group dynamics in the acceptance and promotion of misconceptions, particularly in relation to the identity-protective attitudes that boost polarisation over scientific information. In this study, we successfully test a mediational model between right-wing authoritarianism and pseudoscientific beliefs. First, we carry out a comprehensive literature review on the socio-political background of pseudoscientific beliefs. Second, we conduct two studies (n = 1189 and n = 1097) to confirm our working hypotheses: H1 – intercorrelation between pseudoscientific beliefs, authoritarianism and three axioms (reward for application, religiosity and fate control); H2 – authoritarianism and social axioms fully explain rightists’ proneness to pseudoscience; and H3 – the association between pseudoscience and authoritarianism is partially mediated by social axioms. Finally, we discuss our results in relation to their external validity regarding paranormal and conspiracy beliefs, as well as to their implications for group polarisation and science communication.


Socio-political sources of motivation for pseudoscience
Pseudoscientific claims have been profusely investigated from a cognitive perspective -for example, regarding their close relationship with intuitive cognitive style (Pennycook et al., 2012), causal illusions (Matute et al., 2011), and pseudo-profound bullshit receptivity (Pennycook et al., 2015). In addition, there is a growing corpus of research outcomes on their ideological and political dimensions that has flourished within the 'politically motivated reasoning paradigm' (Kahan, 2016). The effect of minority but influential forms of motivated reasoning elicited by socio-political worldviews (e.g. Palm et al., 2017) is enlightening regarding the recalcitrant nature of certain pseudoscientific beliefs: it makes individuals exposed to information more polarised than nonexposed ones (e.g. Lewandowsky and Oberauer, 2016;Nyhan et al., 2014;Nyhan and Reifler, 2015;Palm et al., 2017), gives rise to backfire effect (e.g. Nyhan et al., 2014;Nyhan and Reifler, 2015), and turns analytical thinking into a polarising factor (e.g. Kahan, 2013). Nevertheless, despite this robust and growing set of research outcomes, group dynamics that underlie this form of politically motivated reasoning are little known.
Recent research on the epidemiology of pseudoscience (e.g.  suggests that identity-protective cognition related to ingroup systems of beliefs, norms and values breaks out when pseudoscientific believers get organised as deviant 'communities of knowledge' (Sloman and Fernbach, 2017). These intergroup struggles would foster perceived threat and uncertainty, which would motivate group identification to boost self-affirmation and uncertainty reduction (Hogg and Wagoner, 2017). Consequently, prior studies have shown that conspiracy theories have a distinctive quadratic relationship with the political spectrum, thus showing associations with general political extremism and dogmatic intolerance elicited by strong beliefs (Van Prooijen et al., 2015) -there is evidence of homogeneity regarding certain cognitive processes between leftwing and right-wing extremists, for example authoritarianism (Conway et al., 2017;Luttig, 2017) and motivated reasoning (Kahan, 2013;Lewandowsky and Oberauer, 2016).
In contrast, the relation between pseudoscience and the political spectrum remains controversial: although there is a confirmed relationship between conservatism and some instances of science denial, such as climate change denial (e.g. Hornsey et al., 2016), other instances, such as genetically modified organism (GMO) opposition, do not show distinctive political associations . Although no prior study has measured the political orientation of pseudoscientific beliefs as a comprehensive construct, it is expected to show a rightward cast by means of greater proneness to conformity, desire to share reality with like-minded others, and ideological echo-chambers among political conservatives (Jost et al., 2018). Further theoretical justification for this expected rightward cast of pseudoscience will be detailed in the following sections.

An authoritarian interpretation of society: The role of conventionalism and intellectual submission
Authoritarianism is a long-established psychological construct, closely related to partisan extremism driven by group-centric affective polarisation (Luttig, 2017). It emerges from fears and uncertainties that give rise to motivated social cognition in which authoritarian attitudes reinforce conformity over social issues (Feldman, 2003), and satisfies epistemic, existential and ideological needs (Jost et al., 2003). Hence, it should be interpreted as a behavioural expression of values that motivates subjects in attaining collective security to the detriment of individual autonomy and critical thinking (Duckitt et al., 2010).
The work of Altemeyer (1981) was momentous due to his robust multidimensional model of authoritarianism, in which the construct is composed of three factors. First, Aggression refers to the disposition to intentionally harm (in psychological, physical or social terms) other individuals or outgroups that are perceived as a threat, accompanied by 'the belief that proper authority approves it or that it will help preserve such authority' (Altemeyer, 1996: 10). Second, Submission refers to the belief that 'proper authorities should be trusted to a great extent and deserve obedience and respect' (Altemeyer, 1996: 9). Accordingly, it boosts the willingness to accept authority's statements and actions without critical assessment. Third, Conventionalism refers to a 'strong acceptance of and commitment to the traditional social norms in one's society' (Altemeyer, 1996: 10). These factors lead individuals to fervently endorse ingroup conventions as social imperatives that must be respected -an inflexible conception of social norms that leads them to reject outgroups' conventions, including their beliefs and values.
Prior research has shown perceived social consensus as a source of motivation for fact assessment. So, individuals tend to accept or reject information depending on whether or not it fits with ingroup values and beliefs (Kahan et al., 2011;, as a way to achieve short-term social benefits (Khanna and Sood, 2017). Accordingly, some cases of denialism show a striking 'consensus-gap' between experts and the public opinion  and perceived group consensus mediates science acceptance on pseudoscientific issues Van der Linden et al., 2015). Therefore, authoritarian predispositions may increase consensus-gap through an increment of radicalism over ingroup identity-related conventions and authorities.
We consider this potential effect of authoritarianism over consensus-gap to be explicable by means of the lay epistemic theory (Kruglanski et al., 2010). Specifically, by means of heightened levels of two of its dimensions: epistemic motivation for non-specific closure (elicited by authoritarian conventionalism) and hyperactive search for epistemic authorities within one's reference group (elicited by authoritarian submission). There is a strong relationship between authoritarianism and need for closure (De Keersmaecker et al., 2017) that motivates subjects to close their minds by 'seizing' on accommodating information and 'freezing' beliefs, thus becoming impervious to adverse data (Kruglanski and Webster, 1996). Group centrism -that is, the degree to which individuals strive to enhance the 'shared-reality' of their collectivity (Kruglanski et al., 2006)involves uniformity pressures, such as denigrating the dissenters or extolling the conformists, in order to achieve group consensus (Kruglanski and Webster, 1991).
This motivated group centrism manifests a preference for opinions that are unlikely to be challenged by significant others, as it would facilitate their esteem and appreciation, as well as the conservation of ingroup ties and social identity. Moreover, group centrism leads individuals to prefer autocratic group structures wherein a centralised authority shields commonly shared opinions (e.g. Pierro et al., 2003). In sum, there are robust reasons to hypothesise that authoritarianism functions as a cognitive framework that hinders the evidential aspect of knowledge formationknown as the 'judgmental unimodel' within lay epistemic theory.

Social axioms as socio-psychological backgrounds that foster the authoritarian dimension of pseudoscience
Social axioms constitute a promising cross-cultural variable to aid understanding of the sociopsychological profile of pseudoscientific believers beyond their distribution in the right-wing/ left-wing political spectrum. The construct is defined by Leung and Bond (2008) as 'generalized beliefs about people, social groups, social institutions, the physical environment, or the spiritual world as well as about categories of events and phenomena in the social world' (p. 10). Social axioms express how society is believed to work through perceived correlational or causal patterns that constitute the basic premises which people endorse and rely upon to make sense of life in society and to guide their actions. In this respect, they differ in several ways from other related constructs such as personality factors (Chen et al., 2006;Leung et al., 2012) and values -such as Hofstede's cultural dimensions, Schwartz's values and cultural worldviews (Leung et al., 2007).
Social axioms are encoded in the form of an assertion about the relationship between two entities or concepts, whereas values describe axiological reasoning and subjective desires; for example, a statement like 'wars are bad' reflects a pacifist value, while statements such as 'powerful people tend to exploit others' are regarded as social axioms due to the specified relationship between the entities, independent of its positive or negative social evaluation (Leung and Bond, 2008). The conceptual differences between social axioms and personality factors are even more explicit, as personality also encompasses attitudes, temperament, values and feelings. In effect, social axioms have shown greater predictive power than personality traits and values in relation to social behaviour (Bond et al., 2004), as they represent a practical guide to interpret societal functioning in a broad range of contexts.
Five social axioms with cross-cultural validity and psychometric soundness have been reported (Leung et al., 2012): Social Cynicism -negative beliefs about human nature, a biased view against some groups of people, mistrust of social institutions and a belief that people disregard ethical means in achieving their ends. For example, 'kind-hearted people are easily bullied' and 'the only way to get ahead is to take advantage of others'.
Reward for Application -effort, careful planning and a belief that the investment of these and other resources will lead to positive social outcomes. Two sample items of this axiom are 'hardworking people are well rewarded' and 'difficult problems can be overcome by hard work and persistence'.
Fate Control -a belief that life events are determined by external forces, but there are some ways for people to influence the impact of these forces. 'Fate determines one's successes and failures' and 'the people whom a person will love in his or her life are determined by fate' are beliefs framed within this axiom.
Social Complexity -a belief that behaviour is inconsistent from situation to situation and that there may be multiple ways of achieving a given outcome. For instance, 'there is usually more than one good way to handle a situation' and 'people may have opposite behaviors on different occasions'.
Religiosity -a belief in the beneficial social functions of religious institutions and practices. Statements such as 'religion helps people make good choices for their lives' and 'religion makes people happier' characterise this social axiom.
Current literature on social axioms has found that three of them (Reward for Application, Religiosity and Fate Control) are positively related to authoritarianism, whereas Social Complexity and Social Cynicism are unrelated (Fasce and Avendaño, 2020). In this study, we expect an analogous pattern regarding pseudoscience.

Religiosity
Religiosity has been already linked to unwarranted beliefs (Singelis et al., 2003) and to social conservatism (Bond et al., 2004). Moreover, the relationship between authoritarian predisposition and normative religious doctrines has been widely documented (e.g. Van Pachterbeke et al., 2011). Moreover, believing in the positive impact of religion among society -for example, over health issues, political decision-making and ethics -involves a lenient attitude towards unwarranted beliefs, as some alternative epistemologies are perceived by these subjects as socially desirable.

Fate control
Fate Control has also been associated with social conservatism, to the endorsement of traditionalism (Bond et al., 2004;Leung et al., 2007), and to unfounded beliefs (Singelis et al., 2003). Hence, this social axiom may be closely linked to the conventionalism dimension of authoritarianism. Fate control could also be related to pseudoscience by means of its existing association with an external locus of control (Chen et al., 2006), and with a conception of facts as shaped by social and political processes (Garrett and Weeks, 2017).

Reward for application
Reward for Application has been positively linked to strengthened obedience towards social norms and authorities (Leung et al., 2007), as it is considered an underlying factor of socially conservative worldviews (Bond et al., 2004). Reward for Application may be promoting pseudoscientific beliefs along a different pathway than Religiosity and Fate Control. People who endorse this social axiom tend to prioritise good social relationships over the defence of potentially conflicting ideas, showing heightened levels of social conformity and uncritical attitude. Thus, they are prone to accommodation as conflict resolution, social desirability, and lack of self-acceptance (Bond et al., 2004;Chen et al., 2006;Singelis et al., 2003). Nevertheless, if Reward for Application works as a background during pseudoscientific belief acquisition, then the mechanism should be strongly susceptible to being suppressed, or reverted to, if short-term incentives change (e.g. Khanna and Sood, 2017).

Social complexity and social cynicism
These two social axioms are expected to be unrelated to pseudoscientific beliefs and authoritarian attitudes. On the one hand, as Social Complexity is positively related to cognitive flexibility (Singelis et al., 2003), problem solving, collaboration, self-direction, and openness to change (Bond et al., 2004), these individuals may react in a more open-minded and self-affirmed way when presented with information that contradicts their belief system, thus promoting flexibility during the assessment of social conventions. On the other hand, there are no theoretical reasons to expect a direct association between Social Cynicism and pseudoscience -although this social axiom may be related to conspiracy theories, as argued in Supplemental Material.

Overview and working hypotheses
We conducted two empirical studies in order to assess the foregoing theoretical framework, structured in the following three working hypotheses: H1 -Pseudoscientific beliefs, religiosity, reward for application, fate control, political orientation and authoritarianism are all positively intercorrelated.
H2 -Religiosity, reward for application, fate control and authoritarianism explain the rightward cast of pseudoscience.
H3 -The association between authoritarianism and pseudoscientific beliefs is mediated by religiosity, reward for application and fate control.

Study 1
Study 1 was designed as an exploratory pilot focused on H1 and H2, and so was conducted to assess the general likelihood of the mediational model displayed in H3. This preliminary study includes validated scales on general pseudoscientific beliefs, the three social axioms hypothesised as related to these beliefs (Reward for Application, Religiosity and Fate Control), and Political Orientation as measured by the right-wing/left-wing axis. Therefore, Study 1 constituted an informative starting point, offering encouraging results that were further replicated and broadened by the thorough confirmatory approach of Study 2.

Sample
We recruited a sample of 1189 Spanish speakers for an online administration of the scales through Google Forms. The respondents were invited to participate using Facebook and Twitter, through forums and groups of pseudoscientific believers. In addition, we counted on the help of science disseminators and sceptic blogs to increase the sample's variability. Given the wide audience of those groups in Spanish-speaking social networks, our sample included participants from Spain and Latin America. In total, 228 (19.2%) were women and 961 (80.8%) were men, with an average age of 39.7 (SD = 10.2). In total, 248 (20.9%) had pre-university education and 941 (79.1%) a university one. Finally, 170 (14.3%) self-describe religious identification and 1019 (85.7%) do not.

Measures
To measure Political Orientation, we included a 10-point Likert-type scale representing the rightwing/left-wing spectrum. To assess pseudoscientific beliefs, we used the 30-item Pseudoscientific Belief Scale (Fasce and Picó, 2019a), a reliable measure (α = .88) 1 that includes pseudo-theory promotion and science denialism as forms of pseudoscience. For the three social axioms included in our model, we used 3 8-item factors extracted from the Social Axioms Survey II (Leung et al., 2012): Reward for Application (α = .88), Fate Control (α = .72), and Religiosity (α = .85). In addition, we included several measures on need to belong and intergroup variables related to pseudoscientific beliefs, to be reported elsewhere.

Results
Sociodemographic variables. There were significant differences in Sex (t = −4.21, d = 0.34, p < .001; more Pseudoscientific Beliefs among women), Education (t = 2.31, d = 0.16, p < .05; more Pseudoscientific Beliefs among subjects with pre-university education), and Religious Identity (t = −8.64, d = 0.76, p < .001; more Pseudoscientific Beliefs among religious subjects). As we did not find significant association between age and pseudoscience, we discarded this variable for further analyses.

H1.
Regarding the association of the variables tested in Study 1, Pseudoscientific Beliefs were positively correlated to right-wing Political Orientation (r = .11, p < .001) and to the three social axioms included: Reward for Application (r = .20, p < .001), Religiosity (r = .33, p < .001) and Fate Control (r = .49, p < .001). These results are displayed in Table 1 and support our first working hypothesis.
H2. In order to test H2, we conducted a hierarchical multiple linear regression with Pseudoscientific Beliefs as the dependent variable. In Model 1, we entered the three sociodemographic variables with significant differences regarding Pseudoscientific Beliefs -namely Sex, Religious Identity, and Education. In Model 2, Reward for Application, Religiosity, and Fate Control were entered as independent variables. Finally, we entered Political Orientation as an independent variable in Model 3, as we wanted to assess its predictive power above social axioms. A multicollinearity test was carried out using variance inflation factor (VIF) and tolerance statistics: all the VIF values were below 1.7 and tolerance statistics were above 0.59. In addition, our data showed independence of errors (Durbin-Watson = 1.92). These results are displayed in Table 2 and support our second working hypothesis. Model 1 (F(3, 1185) = 48.6, p < .001) confirmed the included sociodemographic variables as significant predictors of pseudoscience endorsement, explaining 11% of its variance. Similarly, Model 2 (F(6, 1182) = 89, p < .001) confirmed the three social axioms as significant predictors of Pseudoscientific Beliefs over sociodemographic characteristics, explaining 31% of its variance. In contrast, where Model 3 (F(7, 1181) = 76.4, p < .001) explains the same amount of Pseudoscientific Beliefs' variance, Political Orientation added no predictive power over social axioms and, consequently, was non-significant as a predictor variable. Therefore, social axioms fully explain the positive association between Pseudoscientific Beliefs and right-wing Political Orientation. These results endorse Reward for Application, Religiosity and Fate Control as socio-psychological backgrounds for the dissemination of Pseudoscientific Beliefs.

Study 2
Study 2 was designed as a follow-up to our exploratory results. This second data collection replicated Study 1 and overcame its limitations, by including all the variables and relationships displayed in our working hypotheses -two forms of unwarranted beliefs (pseudoscience and the paranormal), social axioms in full, political orientation and authoritarianism. Consequently, the comprehensive design of Study 2 allowed us to perform a full assessment of our working hypotheses.

Sample
A convenience sample of 1097 Spanish speakers was recruited using the same data collection strategy described in Study 1 for an online fulfilment of the scales. In total, 395 (36%) were women and
H1. As found in Study 1, and as expected by H1, Pseudoscientific Beliefs had positive correlations with Political Orientation (r = .21, p < .001), Conventionalism (r = .28, p < .001), Submission (r = .16, p < .001), Aggression (r = .18, p < .001), and three social axioms: Reward for Application (r = .33, p < .001), Religiosity (r = .36, p < .001), and Fate Control (r = .51, p < .001). These results are displayed in Table 3 and fully support our first working hypothesis. In addition, we found that the complete ASC Scale shows a medium-sized positive correlation with Pseudoscientific Beliefs (r = .31, p < .001). As Pseudoscientific Beliefs show no correlation to Social Cynicism and a very weak, almost non-significant correlation to Social Complexity, we discarded these social axioms for further analyses. These latter results were expected based on our literature review and working hypotheses.
Beliefs' variance, confirming that Political Orientation has no predictive power above authoritarianism and social axioms.
H3. In order to test the hypothesised mediational effects of social axioms on the relationship between authoritarianism and pseudoscientific beliefs, we carried out a series of simple mediational analyses by means of the PROCESS macro (v3.4). Mediation analyses are intended to statistically test hypothesised models in which the relationship between an independent and dependent variable is thought to be influenced by a mediator variable (MacKinnon et al., 2007). Thus, mediator variables explain a causal sequence whereby the independent variable predicts indirectly the outcome on the dependent variable -the so-called 'indirect effect'. In other words, mediation analyses explain how mediating variables intervene in the relationship between the independent and dependent variable.
To verify the present hypothesis (H3), we studied the mediational effects by analysing the indirect effect through bootstrapping (95% confidence intervals; number of bootstrap samples: 5000), revealing that Reward for Application, Religiosity and Fate Control act as partial mediators between authoritarianism (as measured by the whole ASC Scale) and Pseudoscientific Beliefs. All these indirect effects -that is, the total effect minus the direct effect -fully support our third working hypothesis (see Figure 1 for a detailed graphical representation). These results do not suggest a univocal causal pathway between authoritarianism, social axioms and pseudoscience. Instead, based on our literature review, we propose that authoritarian predispositions constitute a cognitive substrate that facilitates the endorsement of certain social axioms; these, in turn, boost the existing association between authoritarianism and unfounded beliefs. Therefore, the observed mediations aid understanding of the primary focus of this article: authoritarianism as a motivational context in which pseudoscience thrives.
Extrapolation to other forms of unwarranted beliefs. We included a validated scale on Paranormal Beliefs to conduct the same analyses as those conducted for Pseudoscientific Beliefs. Interestingly, the pattern of associations, statistical significances and effect sizes of Paranormal Beliefs were very similar to those found in the previous analyses. These results strongly suggest that Paranormal Beliefs are equally related to authoritarianism and partially mediated by the same social axioms as pseudoscience. Hence, the mediational model tested in this study can be rightfully extrapolated to the paranormal, showing that pseudoscience involves strong paranormal content and both groups of believers largely overlap and resemble each other. Results on Paranormal Beliefs and additional remarks on the potential and nuanced extrapolation of these results to conspiracy theories can be found in Supplemental Material.

Discussion
The data reported in this article show that three social axioms partially mediate the existing association between right-wing authoritarianism and pseudoscience. Hence, these results fully endorse our hypotheses, being compatible with a theoretical interpretation framed within lay epistemic theory. Nevertheless, other forms of authoritarianism may be related to specific unwarranted beliefs by means of different social conceptions -such as left-wing authoritarianism (Conway et al., 2017), subtle forms of competitive authoritarianism (Levitsky and Way, 2002) and even liberal authoritarianism (Babones, 2018). For instance, historical examples such as Lysenkoism (Kolchinsky et al., 2017) and current leftist conspiracy theories (e.g. Oliver and Wood, 2014) show that left-wing authoritarianism promotes its own forms of disinformation.

Authoritarianism and social axioms as underlying factors of belief polarisation
Authoritarian predispositions towards social conservatism boosted by social axioms may be explanatory regarding the kind of intergroup struggle that leads to motivated belief polarisation among radical minorities. As recent results strongly suggest that science rejection is mediated by lack of perceived social consensus between experts and the public opinion Van der Linden et al., 2015), the socio-political profile of pseudoscience described in this study may be blocking the perception of expert agreement by reinforcing ingroup conventionalism -although more research is needed to confirm this causal pathway.
Accordingly, recalcitrant unwarranted believers would be reluctant to accept information from the outgroup, particularly due to hyperactive affective anchoring of ingroup membership that heightens the perception of intergroup threats and leads to the rejection of the open marketplace of ideas. These authoritarian individuals would tend to disregard the freedom of expression of the outgroup, avoiding uncertainty by endorsing a monopoly of truth (Hackett et al., 2018), and preferring instead to engage with prototypical, deviant 'truth seekers' (e.g. Franks et al., 2017) -that is, fake experts that hold proper badges of ingroup membership. These authoritarian motives would lead pseudoscientific believers to exploit their analytical thinking in order to rationalise polarisation and partisan science acceptance, thus performing backfire effect. This potential causal chain constitutes a relevant novel research line on the underlying social conceptions that give rise to motivated reasoning and, consequently, block the public acceptance of science.

Implications for social interventions and science communication
The reported association between authoritarianism and pseudoscience suggests that ingroup ostracism and conventionalism may be pushing individuals towards a consensus-gap. Therefore, the most direct intervention would be to offer conditions to improve intergroup contact related to pseudoscientific issues, such as common goals and cooperation (Pettigrew and Tropp, 2008). As a confrontational rhetoric style has been proved to backfire under conditions of motivated reasoning, we must place value on unregulated free speech, conversations that engage diverse viewpoints, and self-disclosure, as these attitudes facilitate mutual understanding (e.g. Turner et al., 2007;Vescio et al., 2003). In fact, this is the typical attitude that can be found among those who hold Social Complexity as an interpretation of the social world -almost unrelated to pseudoscience and authoritarian attitudes.
Another line of interventions may be focused on echo-chambers and partisan media, as they foster social conceptions associated with pseudoscience, such as group bias and, consequently, authoritarian rejection of hostile information. Echo-chambers are often exploited by evidenceresistance groups that effectively promote denialism and pseudo-theories . In general terms, it is important to encourage people to counter the false-consensus effect and harmful intellectual submission by making their voices heard. It would be very helpful to expand the boundaries established by social media algorithms to expose users to a wider spectrum of information from those, including the authoritative voice of scientistswho can participate in the public sphere without risking their credibility (Kotcher et al., 2017). Therefore, it is important for science communication to deploy pedagogical strategies and inoculation messages to cope with disinformation within corrupted information architectures, making the public aware of how fake news, trolling and filter bubbles work Lewandowsky et al., 2012Lewandowsky et al., , 2017.
In addition, even though scientific literacy and critical thinking are negatively correlated to pseudoscience endorsement (Fasce and Picó, 2019b), previous research outcomes consistently concluded that courses that promoted a motivational state of distrust in pseudoscience produced a reduction of those beliefs, whereas general education classes on critical thinking and research methods did not (Dyer and Hall, 2019;Wilson, 2018). 2 So, in light of these results, efficient interventions on pseudoscience endorsement and science communication should include motivational strategies to deal with authoritarianism and counterproductive social conceptions, such as worldview and value affirmation, in order to exploit the existing negative association between trust in science and pseudoscientific beliefs (Fasce and Picó, 2019b;Ståhla and Van Prooijen, 2018).

Limitations
We want to remark on some of the limitations of the reported studies. First, these results must be taken cautiously, particularly in regard to their interpretation in causal terms. These theory-driven correlational results suggest a causal relationship between authoritarian attitudes and pseudoscientific beliefs; however, this potential pathway needs further experimental confirmation, particularly to identify confounders. Second, both samples are composed by a higher number of men, more university educated and more non-religious subjects -even though the samples' variabilities were acceptable enough to include these sociodemographic variables in further analyses. Consequently, these sample asymmetries should be assessed in future studies to confirm that they did not affect the reported results.

Concluding remarks
We have successfully tested a mediational model that characterises pseudoscientific beliefs as related to an authoritarian interpretation of society, in which three social axioms that place great value on unwarranted beliefs play an explanatory role. Hence, the previously reported associations of certain instances of pseudoscience with right-wing ideologies may be explicable by means of this richer socio-political background, related to the lay epistemic theory. Exacerbated levels of authoritarian attitudes may be at the root of motivated reasoning already observed among recalcitrant groups of pseudoscientific believers. As such, some strategic interventions could be beneficial to foster evidence-based behaviours and public acceptance of science.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.