Does religion predict coronavirus conspiracy beliefs? Centrality of religiosity, religious fundamentalism, and COVID-19 conspiracy beliefs.

There has been an increasing interest in the relationship between religion and psychosocial functioning during the COVID-19 pandemic. Interestingly, emerging recent findings suggest that religiousness may have a Janus-face impact on how people cope with the pandemic, leading to both positive and negative social outcomes. In this project, we examine whether two types of religiousness (i.e., centrality of religiosity and religious fundamentalism) are associated with COVID-19 conspiracy beliefs and socially undesirable behavior during the pandemic. We suggest that only the most dogmatic and fundamentalistic type of religiousness could lead to conspiracy beliefs, while centrality of religiosity could be unrelated or even negatively related to this type of thinking. In a series of two studies (N = 361 and N = 394) conducted among Polish Roman Catholics, we demonstrate that religious fundamentalism, unlike centrality of religiosity, is positively related to coronavirus conspiracy beliefs, which, in turn, promote socially maladaptive behavior such as freeriding or non-adherence to safety guidelines.

Introduction

The COVID-19 pandemic has posed an unprecedented threat to both physical and mental health of numerous people worldwide. Scientists have rushed to study social and personal characteristics that could predict, and therefore prevent, the kind of behaviors that could have a detrimental effect on the society during the pandemic, such as the opposition to the COVID-19 mitigation measures or the surge in adherence to conspiracy beliefs regarding the origin and nature of the coronavirus pandemic (e.g., Kowalski et al., 2020; Zajenkowski et al., 2020). Among various factors analyzed, there has been an increasing interest in the role of religion and spirituality. Following that line of research, in this paper, we examine whether religiousness1 in different forms is associated with COVID-19 conspiracy beliefs and socially undesirable behavior during the pandemic.

Religion and COVID-19 pandemic

Religiousness has been long believed to be increasing in times of hardship (e.g., Pargament, 2001; Weber, 1920/1993). Preliminary data accumulated during the current pandemic seems to corroborate this view. For instance, various media outlets informed about the booming sales of religious books, including the Bible, in the first weeks of the global pandemic (e.g., Coyle, 2020). Moreover, researchers found evidence that praying intensified during the current pandemic, including as much as 61% of Poles who reported spending more time on these activities (Bentzen, 2020; Boguszewski et al., 2020). Finally, according to a recent study, nearly three-in-ten Americans (28%) claimed to have stronger personal faith because of the pandemic (Sahgal and Connaughton, 2021). Although we can observe that people increasingly turn toward religion when calamities occur, it remains an open question what kind of influence religiousness might have on their emotionality and behavior during such difficult times.

In general, both previous research and the emerging recent findings suggest that religion may have a Janus-face impact on how people deal with the pandemic (Kranz et al., 2020). On the one hand, a large body of prior evidence indicates a generally positive link between religion and mental health (e.g., Koenig, 2012), with some researchers describing religion as having salutary effects on physical and mental outcomes (Seybold & Hill, 2001). Next, religious individuals are also characterized by higher emotion regulation skills, which may prove particularly useful in coping with pandemic-related stress (Vishkin et al., 2019). Furthermore, previous studies have established a positive association between religiousness and agreeableness (Tsang et al., 2020) or submissiveness (Saroglou et al., 2009), both supposed to predict a higher degree of compliance with pandemic-related restrictions.

On the other hand, the popular narrative in public and scientific discussions is that there exists an unresolved conflict between religion and science (Evans & Evans, 2008), which could result in religious individuals being reluctant to follow the recommendations by public health experts and scientists (Plohl & Musil, 2021). This could also be related to the fact that religious individuals tend to score lower on intelligence and analytic thinking measures (Pennycook et al., 2016; Zuckerman et al., 2020). Further, in terms of mental health, religiousness may also involve maladaptive coping strategies that bring about depressive symptoms and decreased satisfaction with life (e.g., Bjorck & Thurman, 2007). Importantly, Szałachowski and Tuszyńska-Bogucka (2021) have recently found that religion can indeed serve as both a balm and a risk for well-being in times of a pandemic, with certain aspects of religiousness protecting against PTSD symptoms, while other aspects aggravating distress.

Religion and conspiracy beliefs

One particularly negative social outcome of the pandemic is a rise in various types of conspiracy beliefs (Stephens, 2020). Conspiracy beliefs or theories are typically characterized as attempts to explain the origins of significant social events with reference to mystery plots and secretive yet powerful agents working in their own favor at the cost of the society (e.g., Bale, 2007; Douglas et al., 2019). Thus, “most conspiracy beliefs can be framed in terms of beliefs about how a powerful and evil out-group meets in secret, designing a plot that is harmful to one's in-group” (van Prooijen & van Lange, 2014, pp. 238–239; emphasis added; see also Cichocka, Marchlewska, Golec de Zavala, & Olechowski, 2016).

Importantly, studies on conspiracy thinking are mainly focused on identifying an array of personality characteristics that are linked to a conviction that others are secretly conspiring against us. In other words, instead of analyzing whether a particular conspiracy theory is true or false, social scientists usually try to understand the role of adopting such beliefs in dealing with distress (e.g., Marchlewska et al., 2021), self-evaluation problems (Cichocka, Marchlewska, & Golec de Zavala, 2016) uncertainty (Marchlewska et al., 2018), or other types of psychological weaknesses (for a review see Biddlestone et al., 2021). They also explore potential consequences of adopting conspiratorial explanations (e.g., the relationship between vaccination conspiracy beliefs and lower support for voluntary vaccination policy; Cislak et al., 2021). In a similar vein, recent research has shown that the endorsement of conspiracy theories regarding COVID-19 is negatively related to the adherence to epidemiological safety guidelines (Kowalski et al., 2020). It also reduces social distancing over time (Bierwiaczonek et al., 2020) and is positively associated with resistance to other preventive actions, including vaccinations (Romer & Jamieson, 2020).

Somewhat contrary to the popular belief, however, the relationship between religion and conspiracy thinking is quite complex and not yet fully grasped (Jasinskaja-Lahti & Jetten, 2019). For example, some researchers draw close comparisons between conspiracy theories and institutionalized religions or even suggest that conspiracy beliefs are of quasi-religious character (Franks et al., 2013). From the cognitive perspective, there is, indeed, some significant overlap between conspiracy and a religious mindset. This includes, for instance, a tendency to detect agency, communication rituals, or a minimally counterintuitive nature of both religious and conspiracy beliefs (Franks et al., 2013). What is more, supernatural and conspiracy beliefs at times may co-occur, as in the case of the so-called “conspirituality” - a term coined by Ward and Voas (2011). Conspirituality refers to a politico-spiritual philosophy, characteristic of the New Age movement, in which beliefs in secret powerful agents are combined with the conviction that humanity is undergoing a shift in consciousness (Ward & Voas, 2011). On the other hand, some empirical investigations do report no significant associations between conspiracy theories and various aspects of religiousness (e.g., Jasinskaja-Lahti & Jetten, 2019; Ladini, 2021). For example, Marchlewska et al. (2019) found that it was defensive in-group identity (e.g., Catholic collective narcissism), and not religiosity per se, that predicted gender conspiracy beliefs (i.e., a conviction that gender studies represent an ideology designed to secretly harm traditional values and social arrangements). Furthermore, it has been suggested that religion can also serve as a buffer, having a preventive role against certain types of conspiracy beliefs that explicitly violate religious dogma (e.g., Ladini, 2021; Norman, 2021).

Religiosity and religious fundamentalism

To address these conflicting findings and predictions regarding conspiracy beliefs and religion, we suggest that general religiosity should be differentiated from more dogmatic and orthodox forms of religiosity, such as religious fundamentalism. In this view, only the most dogmatic and fundamentalistic types of religiousness would lead to conspiracy, while general religiosity would be unrelated, or even negatively related, to this type of thinking due to its positive link with agreeableness and prosociality (Tsang et al., 2020). A large and growing body of literature has investigated the specificity of religious fundamentalism (i.e., an unwavering belief in a set of absolute religious dogmas as expressed literally in sacred scriptures; e.g., Zhong et al., 2017). It has been demonstrated that religious fundamentalists are particularly prejudiced and prone to aggression toward out-group members (e.g., Vincent et al., 2011), have lower scientific literacy (Sherkat, 2011), and generalized negative attitude toward secular education (Sherkat & Darnell, 1999).

Importantly, similar effects have not been consistently observed for more general measures of religiousness, including the Centrality of Religiosity Scale, which describes the subjective importance of religious meanings within one's personality structure (Huber & Huber, 2012). For example, Yendell and Huber (2020) found that non-Muslim Swiss adults who scored higher on religion's centrality, unlike religious fundamentalists, had more positive attitudes toward Islam. Furthermore, a recent investigation showed that while high levels of religiosity are inconsistent with high valuation of science in some cultures, religiosity and science might be entirely compatible in other countries, such as Iran (Payir et al., 2021). Finally, new evidence suggests that, on average, religious people not only view science and religion as equally instrumental sources of knowledge, they also perceive less conflict between science and religion than non-religious people (Jackson et al., 2020).

Overall, this state of knowledge clearly highlights that studies on religion's link with the social response to the coronavirus pandemic should take into account both the general importance of religiosity in a personal system and the particular type of religious attitude that expresses itself in strong and exceptionless endorsement of an infallible set of beliefs.

Current research

Although there exist some findings linking religion with conspiracy beliefs (e.g., Franks et al., 2013; Galliford & Furnham, 2017; Ward & Voas, 2011), the relationship between these two is not evident and two opposing predictions could be made regarding the role of religion. In the current research, we decided to investigate the relationships between centrality of religiosity, religious fundamentalism and COVID-19 conspiracy beliefs. In Study 1, we recruited a nationwide sample of Polish adults to establish general links between the two types of religiousness and conspiracy beliefs regarding the coronavirus pandemic. Study 2 was intended to both conceptually replicate the findings from Study 1 and to extend the scope of the previous research by including measures of socially undesirable behavior, such as freeriding or non-adherence to COVID-19 prevention guidelines. It was hypothesized that religious fundamentalism, but not centrality of religiosity, would be positively associated with COVID-19 conspiracy beliefs (H1; Study 1 and Study 2). Additionally, we assumed that religious fundamentalism (but not centrality of religiosity) would be positively associated with freeriding during the COVID-19 pandemic (H2: Study 2) and negatively associated with compliance with safety and self-isolation guidelines (H3: Study 2). In such a way, we aimed to check whether COVID-19 conspiracy beliefs might account for the relationship between religious fundamentalism and socially undesirable behavior during the pandemic.

Study 1

Methods

Participants and procedure

This study involved a nationwide sample of Polish adults. A total number of 432 participants completed the survey: 183 women (coded as 0), 249 men (coded as 1), aged 18–84 (M = 48.18, SD = 16.34). Because we were interested in examining the links between conspiracy beliefs regarding the coronavirus pandemic and religious fundamentalism versus centrality of religiosity among Catholics, we excluded anyone who reported their religious affiliation as other than Catholic from further analyses (n = 71). Thus, the final sample consisted of 361 respondents (157 women; 204 men) between the ages of 18 and 84 (M = 48.25, SD = 16.33). Data was collected via an internet questionnaire (CAWI), by an external research company – Pollster Institute – a Polish online research platform that has been used in academic studies before (e.g., Kowalski et al., 2020).

Participants filled out measures of religious fundamentalism, centrality of religiosity, conspiracy beliefs regarding the coronavirus pandemic, as well as demographics among other variables.2

Measures

Religious Fundamentalism was measured using the Religious Fundamentalism Scale (Altemeyer & Hunsberger, 2004; 12-item Polish adaptation, Besta & Błażek, 2007). Respondents were asked to what extent they agreed or disagreed with statements such as “God has given mankind a complete, unfailing guide to happiness and salvation, which must be totally followed” and “The basic cause of evil in this world is Satan, who is still constantly and ferociously fighting against God” using a scale from 1 = very strongly disagree to 8 = very strongly agree, α = 0.89, M = 4.30, SD = 1.27.

Centrality of Religiosity was measured using the Centrality of Religiosity Scale by Huber and Huber (2012; Polish adaptation by Zarzycka, 2011) consisting of 5 items referring to the intellect dimension (i.e., “How often do you think about religious issues?”, on a scale from 1 = never to 5 = very often), the ideology dimension (i.e., “To what extent do you believe that God or something divine exists?”, on a scale from 1 = not at all to 5 = very much so), the experience dimension (i.e., “How often do you experience situations in which you have the feeling that God or something divine intervenes in your life?”, on a scale from 1 = never to 5 = very often), the public practice dimension (i.e., “How often do you take part in religious services?”, on a scale from 1 = never to 7 = more than once a week), and the private practice dimension (i.e., “How often do you pray?”, on a scale from 1 = never to 9 = several times a day), α = 0.86, M = 2.77, SD = 0.84. For dimensions of public and private religious practice the response format was based on objective frequencies of a given activity. The objective frequencies were then recoded to subjective frequencies ranging from 1 to 5, according to the original instructions by Huber and Huber (2012).

Conspiracy Beliefs were measured with an 11-item questionnaire, based on a scale previously used by Kowalski et al. (2020). We measured various types of conspiracy beliefs regarding the coronavirus pandemic such as “Coronavirus was created by ecologists to reduce population number and help the environment” or “Coronavirus was created by pharmaceutical organizations”. Participants responded on a scale from 1 = definitely disagree to 5 = definitely agree, α = 0.95, M = 2.45, SD = 1.03. For some initial psychometric evaluation of the scale, please see the Supplementary material (available online).

Results

Zero-order correlations

First, we computed correlations between all of the variables (see Table 1 ). Conspiracy beliefs were significantly and positively correlated with religious fundamentalism but non-significantly with centrality of religiosity. Religious fundamentalism was highly and positively correlated with centrality of religiosity.

Table 1

Zero-order correlations between key variables (Study 1).

Variables	1	2	3
1. Religious fundamentalism	–
2. Centrality of religiosity	0.69⁎⁎⁎	–
3. Coronavirus conspiracy beliefs	0.18⁎⁎⁎	0.06	–

Note. N = 361.

p < .001.

Regression analysis

To test which form of religiousness predicts beliefs in conspiracy theories regarding the coronavirus pandemic, we included religious fundamentalism and centrality of religiosity as predictors in a multiple regression analysis. We also included age and gender as covariates (Table 2 ). The results showed a significant and positive effect of religious fundamentalism (but not centrality of religiosity) on conspiracy beliefs. We also found significant negative effects of gender and age on conspiracy beliefs. Please note that in the Supplementary material we also report all subsequent analyses without the inclusion of gender and age.

Table 2

The effects of religious fundamentalism and centrality of religiosity on coronavirus conspiracy beliefs (Study 1).

Variables	β	B (SE)	95% CI
Intercept		2.57 (0.26)	[2.05, 3.09]
Religious fundamentalism	0.24⁎⁎⁎	0.20 (0.06)	[0.08, 0.31]
Centrality of religiosity	−0.11	−0.14 (0.09)	[−0.31, 0.03]
Gender (0 = female; 1 = male)	−0.13⁎⁎	−0.28 (0.11)	[−0.49, −0.07]
Age	−0.14⁎⁎	−0.01 (0.003)	[−0.01, −0.002]
R2	0.08
F	F(4, 356) = 8.009⁎⁎⁎

p < .001.

p < .01.

Discussion

Study 1 showed that religious fundamentalism, but not centrality of religiosity, was positively linked to coronavirus conspiracy beliefs. In such a way, Study 1 confirmed our basic prediction that not all types of religiosity are linked to adopting conspiracy explanations. These results suggest that only dogmatic and orthodox forms of religiosity, such as religious fundamentalism, may have a potential to increase beliefs in secret plots by powerful and malevolent groups. This is also in line with previous research showing positive links between religious fundamentalism and out-group hostility (Vincent et al., 2011), since conspiracy theories have been likewise associated with angry and hostile reactions toward other people or groups of people (van Prooijen & van Vugt, 2018).

In Study 2, we aimed to replicate the pattern of results obtained in Study 1 and to explore the destructive correlates of adopting coronavirus conspiracy beliefs. Specifically, we examined whether the endorsement of coronavirus conspiracy beliefs might be responsible for the relationship between religious fundamentalism and risky behavior related to the coronavirus pandemic. In other words, we assumed that religious fundamentalism (but not centrality of religiosity) should be linked to an increase in freeriding behaviors during the COVID-19 pandemic (e.g., taking advantage of the fact that the streets were empty to carry out personal plans) and a decrease in adherence to safety and self-isolation guidelines (e.g., social distancing). We also hypothesized that these relationships would be accounted for by coronavirus conspiracy beliefs. Moreover, in order to target Polish religious believers more specifically, in Study 2 we used a different measure of religious fundamentalism, the one that corresponds more closely with the context of radical Catholicism (Czarnek et al., 2017). In terms of our study participants, we decided to focus on the sample of young adults because previous research has shown that individuals in this transformative period of life might be particularly prone to experience uncertainty and loss of control associated with the COVID-19 pandemic (e.g., Shanahan et al., 2020).

Study 2

Data for Study 2 was obtained in a 2020 online survey, conducted by the Ariadna research panel, which has been frequently used in academic studies (e.g., Marchlewska et al., 2019). The sample consisted of 558 respondents (272 women, coded as 0; 286 men, coded as 1) between the ages of 18 and 26 (M = 23.30, SD = 2.12). Same as in Study 1, we were interested in examining the links between conspiracy beliefs and religious fundamentalism versus centrality of religiosity among Catholics. Thus, we excluded from the analyses anyone who reported their religious affiliation as other than Catholic (n = 164). The final sample consisted of 394 respondents (195 women; 199 men) between the ages of 18 and 26 (M = 23.46, SD = 2.05). Participants filled out measures of religious fundamentalism, centrality of religiosity, coronavirus conspiracy beliefs, freeriding during the pandemic, adherence to safety and self-isolation measures, as well as demographics among other variables.

Religious Fundamentalism was measured using the Political Beliefs Questionnaire (Czarnek et al., 2017), in which 6 items measured religious fundamentalism. Respondents were asked to what extent they agreed or disagreed with statements such as “Christian values should be particularly protected in Poland” and “Poland should be a more Catholic country.” using a scale from 1 = definitely disagree to 5 = definitely agree, α = 0.86, M = 2.97, SD = 0.89. Considering the kind of questions within the scale, it might be also conceptualized as a measure of radical religious nationalism, or more specifically as a measure of Polish Catholic nationalism (see Al-Kire et al., 2021).

Centrality of Religiosity as in Study 1, was measured using the five-item Centrality of Religiosity Scale (Huber & Huber, 2012; Polish adaptation by Zarzycka, 2011), α = 0.84, M = 2.95, SD = 0.96.

Conspiracy Beliefs were measured in a similar way to that in Study 1, based on a scale previously used by Kowalski et al. (2020). Participants responded on a scale from 1 = definitely disagree to 7 = definitely agree, α = 0.96, M = 3.25, SD = 1.49.

Freeriding During the Pandemic was measured by a 3-item scale based on Molenda et al.'s (2020) negative behaviors during the pandemic scale. Participants were asked to what extent they agreed or disagreed with the following statements: “Taking advantage of the fact that the streets were empty, I willingly went out for walks.”, “I was able to take the coronavirus protection recommendations with ease because other people followed them.”, “I met with friends often and willingly” using a scale from 1 = definitely disagree to 7 = definitely agree, α = 0.78, M = 3.25, SD = 1.41.

Adherence to Safety and Self-Isolation Guidelines was measured with a 4-item scale created for the purposes of this study: “I try to my keep distance from others when I'm outside”, “I go shopping wearing a mask”, “I participate in big events (for example weddings, big house parties)”; reverse coded item, “I participate in mass events that are against the law”; reverse coded item. Participants were asked to determine whether the statement was true for them, using a scale from 1 = definitely false to 7 = definitely true, α = 0.73, M = 5.08, SD = 1.24.

We computed correlations between all of the variables (see Table 3 ). Religious fundamentalism was significantly and positively associated with centrality of religiosity, freeriding during the pandemic and conspiracy beliefs. Centrality of religiosity was significantly and negatively correlated only with conspiracy beliefs. Coronavirus conspiracy beliefs were positively correlated with freeriding and negatively with adherence to safety and self-isolation guidelines. Freeriding and adherence to safety and self-isolation were negatively related to each other.

Table 3

Zero-order correlations between key variables (Study 2).

Variables	1	2	3	4	5
1. Religious fundamentalism	–
2. Centrality of religiosity	0.42⁎⁎⁎	–
3. Freeriding during the pandemic	0.10⁎	−0.05	–
4. Adherence to safety and self-isolation guidelines	−0.09	0.07	−0.47⁎⁎⁎	–
5. Coronavirus conspiracy beliefs	0.20⁎⁎⁎	−0.13⁎⁎	0.49⁎⁎⁎	−0.56⁎⁎⁎	–

Note. N = 394.

p < .001.

p < .01.

p < .05.

Regression analyses

First, we computed a hierarchical regression analysis to investigate the effects of religious fundamentalism and centrality of religiosity on conspiracy beliefs regarding the coronavirus pandemic (Table 4 ). We also included demographic variables, such as gender and age. Similarly to Study 1, we found a significant positive effect of religious fundamentalism on conspiracy beliefs. This time, however, the effect of centrality of religiosity on conspiracy beliefs was also significant and negative. The effects of gender and age were non-significant.

Table 4

The effects of religious fundamentalism and centrality of religiosity on coronavirus conspiracy beliefs (Study 2).

Variables	β	B (SE)	95% CI
Intercept		3.20 (0.89)	[1.46, 4.94]
Religious fundamentalism	0.31⁎⁎⁎	0.52 (0.09)	[0.35, 0.70]
Centrality of religiosity	−0.26⁎⁎⁎	−0.41 (0.08)	[−0.57, −0.25]
Gender (0 = female; 1 = male)	0.03	0.09 (0.14)	[−0.20, 0.37]
Age	−0.02	−0.01 (0.03)	[−0.08, 0.05]
R2	0.10
F	F(4, 389) = 11.040⁎⁎⁎

p < .001.

Second, we computed another hierarchical regression analysis but this time we investigated the effects of religious fundamentalism, centrality of religiosity and conspiracy beliefs on freeriding behaviors during the pandemic (see Table 5 ). In the first step, we introduced religious fundamentalism and centrality of religiosity, together with the demographic variables. The effect of religious fundamentalism was significant and positive. Centrality of religiosity was a significant and negative predictor of freeriding during the pandemic. Age also resulted a significant and negative predictor of freeriding during the pandemic (but not gender). In the second step, we introduced coronavirus conspiracy beliefs. We found that conspiracy beliefs were a significant and positive predictor of freeriding during the pandemic. Importantly, after introducing conspiracy beliefs into the regression model, the effects of religious fundamentalism and centrality of religiosity became non-significant. We did not find a significant effect of gender but the effect of age remained significant, albeit weak.

Table 5

The effects of religious fundamentalism, centrality of religiosity and coronavirus conspiracy beliefs on freeriding during the pandemic.

Variables	Model 1			Model 2
	β	B (SE)	95% CI	β	B (SE)	95% CI
Intercept		4.71 (0.87)	[3.01, 6.42]		3.24 (0.78)	[1.71, 4.77]
Religious fundamentalism	0.13⁎	0.21 (0.09)	[0.04, 0.39]	−0.02	−0.03 (0.08)	[−0.18, 0.13]
Centrality of religiosity	−0.11⁎	−0.16 (0.08)	[−0.32, −0.001]	0.02	0.03 (0.07)	[−0.11, 0.17]
Gender (0 = female; 1 = male)	0.04	0.13 (0.14)	[−0.15, 0.41]	0.03	0.09 (0.12)	[−0.16, 0.33]
Age	−0.10⁎	−0.07 (0.03)	[−0.14, −0.005]	−0.09⁎	−0.07 (0.03)	[−0.12, −0.01]
Coronavirus conspiracy beliefs				0.49⁎⁎⁎	0.46 (0.04)	[0.37, 0.55]
Adjusted R2	0.03			0.24
F	F(4, 389) = 3.499⁎⁎			F(5, 388) = 25.537⁎⁎⁎
ΔR2				0.21
ΔF				F(1, 388) = 109.779⁎⁎⁎

p < .001.

p < .01.

p < .05.

In order to perform a full test of our hypotheses, we conducted a mediation analysis using model 4 with one mediator in Process (Hayes, 2018). The analysis displayed in Fig. 1 examined whether conspiracy beliefs mediated the path between religious fundamentalism and freeriding during the pandemic. As covariates we used the centrality of religiosity, gender, and age. The indirect effect of religious fundamentalism on freeriding during the pandemic via conspiracy beliefs of 0.240 was significant, with a bootstrapped 95% bias-corrected confidence interval of 0.143 to 0.348.

Fig. 1

Indirect effect of religious fundamentalism on freeriding during the pandemic via coronavirus conspiracy beliefs.

Note: Entries are unstandardized coefficients. Dotted line indicates total effect (not controlling for the third variable). *p < .05. **p < .01. ***p < .001.

Then, we conducted a second mediation analysis (model 4) to test whether conspiracy beliefs mediated the path between centrality of religiosity and freeriding during the pandemic (see Fig. 2 ). As covariates we used religious fundamentalism, gender and age. The indirect effect of centrality of religiosity on freeriding during the pandemic via conspiracy beliefs of −0.188, was significant with bootstrapped 95% bias-corrected confidence intervals of −0.284 to −0.103.

Fig. 2

Indirect effect of centrality of religiosity on freeriding during the pandemic via coronavirus conspiracy beliefs.

Note: Entries are unstandardized coefficients. Dotted line indicates total effect (not controlling for the third variable). *p < .05. **p < .01. ***p < .001.

Finally, we conducted another hierarchical regression analysis to investigate the effects of religious fundamentalism, centrality of religiosity, and conspiracy beliefs on an adherence to safety and self-isolation guidelines (see Table 6 ). First, we introduced religious fundamentalism, centrality of religiosity and the demographic variables. The effect of religious fundamentalism was significant and negative, while the effect of centrality of religiosity was significant and positive. The effect of age was non-significant, but gender was a significant and negative predictor of the adherence to safety and self-isolation guidelines. In the second step, we introduced coronavirus conspiracy beliefs. We found that conspiracy beliefs were a significant and negative predictor of the dependent variable. Moreover, the effects of religious fundamentalism and centrality of religiosity were no longer significant. Gender remained a significant and negative predictor, while the effect of age was still non-significant.

Table 6

The effects of religious fundamentalism, centrality of religiosity and coronavirus conspiracy beliefs on adherence to safety and self-isolation guidelines.

Variables	Model 1			Model 2
	β	B (SE)	95% CI	β	B (SE)	95% CI
Intercept		5.16 (0.76)	[3.66, 6.65]		6.67 (0.65)	[5.40, 7.94]
Religious fundamentalism	−0.13⁎	−0.18 (0.08)	[−0.34, −0.03]	0.04	0.06 (0.07)	[−0.07, 0.19]
Centrality of religiosity	0.12⁎	0.16 (0.07)	[0.02, 0.30]	−0.03	−0.03 (0.06)	[−0.15, 0.09]
Gender (0 = female; 1 = male)	−0.12⁎	−0.30 (0.12)	[−0.54, −0.05]	−0.10⁎	−0.26 (0.10)	[−0.46, −0.05]
Age	0.01	0.01 (0.03)	[−0.05, 0.06]	−0.001	−0.001 (0.02)	[−0.05, 0.05]
Coronavirus conspiracy beliefs				−0.57⁎⁎⁎	−0.47 (0.04)	[−0.54, −0.40]
Adjusted R2	0.03			0.32
F	F(4, 389) = 3.799⁎⁎			F(5, 388) = 38.008⁎⁎⁎
ΔR2				0.29
ΔF				F(1, 388) = 168.309⁎⁎⁎

p < .001.

p < .01.

p < .05.

We conducted a mediation analysis using model 4 with one mediator in Process (Hayes, 2018) to examine whether conspiracy beliefs mediated the path between religious fundamentalism and adherence to safety and self-isolation guidelines (see Fig. 3 ). As covariates we used the measure of centrality of religiosity, gender, and age. The indirect effect of conspiracy beliefs on adherence to safety and self-isolation guidelines of −0.246 was significant, with a bootstrapped 95% bias-corrected confidence interval of −0.342 to −0.154.

Fig. 3

Indirect effect of religious fundamentalism on adherence to safety and self-isolation guidelines via coronavirus conspiracy beliefs.

Note: Entries are unstandardized coefficients. Dotted line indicates total effect (not controlling for the third variable). *p < .05. **p < .01. ***p < .001.

Then, we conducted a second mediation analysis (model 4) to test whether conspiracy beliefs mediated the path between centrality of religiosity and adherence to safety and self-isolation guidelines (see Fig. 4 ). As covariates we used religious fundamentalism, gender, and age. The indirect effect of centrality of religiosity on adherence to safety and self-isolation guidelines via conspiracy beliefs of 0.193, was significant with bootstrapped 95% bias-corrected confidence intervals of 0.110 to 0.282.

Fig. 4

Indirect effect of centrality of religiosity on adherence to safety and self-isolation guidelines via coronavirus conspiracy beliefs.

Note: Entries are unstandardized coefficients. Dotted line indicates total effect (not controlling for the third variable). *p < .05. **p < .01. ***p < .001.

In Study 2, we replicated the results obtained in Study 1, suggesting that it is indeed religious fundamentalism, rather than centrality of religiosity, that positively predicts coronavirus conspiracy beliefs. Study 2 additionally revealed that religious fundamentalism was a positive predictor of risky behaviors during the coronavirus pandemic (i.e., higher freeriding and lower adherence to safety and self-isolation guidelines) and that these relationships were accounted for by coronavirus conspiracy beliefs. Moreover, the results of Study 2 also showed that centrality of religiosity predicted lower freeriding during the pandemic and higher adherence to safety and self-isolation guidelines. These relationships were also accounted for by coronavirus conspiracy beliefs.

These conclusions are in line with previous studies (e.g., Vincent et al., 2011; Yendell & Huber, 2020) showing the opposite effects of religious fundamentalism versus centrality of religiosity on intergroup relations, but also intragroup behaviors. Those high (vs. low) in religious fundamentalism seem to engage in maladaptive behaviors both from the perspective of inter- (i.e., conspiracy beliefs) and intragroup (i.e., risky behaviors during the pandemic) relations. This, however, is not the case among those high (vs. low) in centrality of religiosity who reject conspiracy beliefs and, thus, are more prone to engage in protective behaviors during the pandemic.

General discussion

In the current research, we investigated the relationships between the two types of religiousness (centrality & fundamentalism) and the COVID-19 social response, including conspiracy beliefs and undesirable social behavior. We confirmed our hypothesis that religious fundamentalism is positively associated with coronavirus conspiracy beliefs. This effect can be explained within the framework of motivational underpinnings of religion. It has been argued that people are attracted to religions because they provide opportunities to fulfill some fundamental human needs, such as the need to know, the need to belong, or the need for personal significance (Szumowska et al., 2020). Although these proclivities are shared by all individuals, people also typically differ in strength and extent of their needs. For instance, we all have a different scope of preference for firm, stable, and definite knowledge on certain topics, a phenomenon known as the need for cognitive closure (Webster & Kruglanski, 1994). Interestingly, religious fundamentalism understood as “a distinctive attitude of certainty as to the ultimate truth of one's religious beliefs” (Kossowska et al., 2018, p. 1) shares some core features with the need for cognitive closure and previous empirical work has established a positive association between both constructs (e.g., Brandt & Reyna, 2010). On the other hand, the need for cognitive closure has been also found to predict the endorsement of conspiracy theories (Marchlewska et al., 2018). Specifically, those high in the need for cognitive closure were found to seize on conspiratorial explanations under conditions of uncertainty. As conspiracy beliefs represent a closed, certain, and structured mode of thinking, they have a potential to provide closure (Marchlewska et al., 2018), which seems of great importance in times of the coronavirus. This would suggest that the need for cognitive closure may boost religious fundamentalism and further lead to conspiracy beliefs. This issue, however, requires further empirical investigation.

The impact of religious fundamentalism on conspiracy beliefs should be compared and confronted with the association between centrality of religiosity and conspiracy beliefs. In our research, we found that while centrality of religiosity was strongly and positively associated with religious fundamentalism across both studies, it was also either unrelated (Study 1) or negatively related (Study 2) to coronavirus conspiracy beliefs. The centrality of religiosity refers to the salience of religious meanings in one's personality; however, it does not tap into any specific contents of religion (Huber & Huber, 2012). In other words, while centrality describes whether religion occupies a central or a marginal role in the system of personal constructs, it does not directly determine the substance of religious beliefs endorsed by an individual (see Zarzycka et al., 2020). Therefore, depending on a specific case, high centrality of religiosity could be sometimes positively related to antagonistic and hostile beliefs (e.g., Hannover et al., 2018), but it also generally tends to associate with more prosocial and open-minded religious convictions (e.g., Łowicki & Zajenkowski, 2020; Yendell & Huber, 2020). This helps to explain why we observe that, on average, centrality of religiosity is a negative predictor of coronavirus conspiracy beliefs and socially undesirable behavior during the pandemic.

Importantly, Study 2 showed that conspiracy beliefs accounted for the link between a. high fundamentalism and b. low centrality of religiosity and socially undesirable behaviors during the pandemic (i.e., higher freeriding and lower adherence to epidemiological safety guidelines). These results are in line with previous findings linking conspiracy beliefs to negative societal outcomes (e.g., Kowalski et al., 2020). Future research would do well to further explore these relationships and investigate, for example, whether interventions aimed at attenuating religious fundamentalism may lower conspiracy beliefs and, as a result, boost positive social behaviors during a pandemic.

In terms of the social relevance of our findings, we suggest that contrary to some people's beliefs (especially non-religious people), not every religious person has to be at odds with science and scientific explanations of extraordinary events such as a pandemic (e.g., Jackson et al., 2020; McPhetres et al., 2021). As our results are in line with previous findings regarding negative outcomes of religious fundamentalism (e.g., prejudice toward outgroups; Vincent et al., 2011), it is worth considering promoting less dogmatic and more healthy forms of religiosity among religious individuals in future interventions.

Limitations and future directions

Although our research sheds a new light on the role of two types of religiousness, coronavirus conspiracy beliefs and behavior during pandemics, it is not without limitations. First of all, both studies were cross-sectional and involved correlational designs, thereby limiting causal inferences. It is worth noticing that despite showing several significant indirect effects, our results do not provide evidence for the causal model because of the abovementioned cross-sectional nature of the studies. Future studies, employing experimental or longitudinal designs, are needed to establish the causality of the observed relationships. We also used a novel coronavirus conspiracy beliefs scale. Although this measure was used in previous research (Kowalski et al., 2020), its psychometric properties were not thoroughly investigated. Thus, future studies are crucial to fully examine the scale's validity, ideally on representative samples drawn from different populations. Furthermore, our research was conducted only among Polish participants, making our results less generalizable. While this investigation focused on Roman Catholics in Poland, our findings should be replicated in other cultural contexts, including non-WEIRD samples. Moreover, potentially fertile ground for future research would also be to investigate the relations between different forms of religiousness and not only acceptance but also the mere rejection of conspiracy explanations. In fact, there are many mysterious world events that may be difficult to explain in different than conspiratorial terms (e.g., Malaysia Airlines Flight MH370 plane disappearance). The fact that one does not accept a conspiracy explanation (i.e., scores low on particular scale measuring conspiracy beliefs) does not clarify whether they reject conspiracy theory (“I am sure that there was no conspiracy”) or simply accepts uncertainty and a lack of knowledge related to a particular event (“I do not know what exactly happened”). These two answers are different, but scales used to measure conspiracy beliefs do not seem to address this issue precisely. Thus, further empirical investigation is needed to better understand the differences between those who accept uncertainty versus reject conspiracy explanations and seize on non-conspiratorial alternatives that are not necessarily based on facts. Finally, recent research has found that the effects of personal characteristics are less pronounced in strong situations such as the current pandemic (Zajenkowski et al., 2020). Future research on religion's role during the pandemic should, therefore, include situational perceptions as well as other types of social and cognitive factors that may influence the beliefs and the behavior of individuals in the wake of the COVID-19 pandemic.

CRediT authorship contribution statement

Paweł Łowicki: Conceptualization, Methodology, Writing – original draft, Writing – review & editing. Marta Marchlewska: Conceptualization, Methodology, Writing – original draft, Writing – review & editing, Funding acquisition. Zuzanna Molenda: Methodology, Formal analysis. Adam Karakula: Formal analysis, Writing – original draft. Dagmara Szczepańska: Writing – review & editing.

Declaration of competing interest

The authors declare that they have no conflict of interest.