Psychology of conspiracy mentality

Conspiracy theories offer alternative explanations for significant events, suggesting the presence of covert schemes orchestrated by influential and malevolent factions [13]. These theories frequently challenge conventional narratives, positing that official accounts of events are misleading, concealing hidden truths [14]. COVID-19 fosters the creation of conspiracy theories, as such theories tend to emerge during threatening moments of crisis that breed uncertainty [15]. Some of the most prevalent theories include that 5G technology is responsible for the transmission of the virus, that the virus is a deliberate creation as a biological weapon originating from China, and that Bill Gates is exploiting the pandemic to establish global control through mandatory vaccination and surveillance initiatives. Although various conspiracy theories might seem to be irrelevant in their content, they are rooted in varying degrees of mistrust or antipathy against higher-status groups, officials, or established mainstream accounts.

Belief in conspiracy theories is prevalent among the general public, particularly during societal crisis situations such as a pandemic [10, 16]. A social media analysis of COVID-19-related infodemic around the globe revealed that 7.8% of the 2,311 reports in 25 languages from 87 countries were identified as conspiracy theories [17]. In a recent survey regarding COVID-19 conspiracy beliefs, 35% of the Turkish population strongly agreed that COVID-19 was created in a laboratory as a biological weapon, while 28% strongly agreed that events like COVID-19 were caused by a small group with secret goals and 26% strongly agreed that scientists intentionally deceived the public regarding the virus [18]. Moreover, researchers consistently find that endorsing one particular conspiracy statement increases the likelihood of believing seemly unrelated and even mutually incompatible ones [19–22]. These findings have led scholars to argue that a broader worldview, often referred to as a “conspiracy mentality”, may underpin conspiracy beliefs, and this mentality tends to remain stable over time [8, 23–26]. This general worldview, marked by a perception of the world as rife with conspiracies, can contribute to the acceptance of conspiracy theories pertaining to specific events, such as those related to COVID-19.

Conspiracy mentality has important social and health-related consequences [27] including reduced engagement with mainstream politics, climate change initiatives, vaccination programs, and conventional medical treatments [5]. Furthermore, it has been recognized as a potential hindrance to constructive public responses during the pandemic [10]. Drawing this research together, we hypothesize that conspiracy mentality might negatively impact adherence to COVID-19 mitigation measures, and we focus on the emotional and cognitive mechanisms involved.

Affective intelligence theory

Considering the world has been repeatedly hit by pandemics such as Spanish flu, SARS, and swine flu, it is evident that negative emotions play a crucial role in shaping individuals’ responses to crises [28]. Affective intelligence theory (AIT) posits that emotions serve as an adaptive system that significantly influences our cognition and subsequent behaviors [29]. Emotions are not merely by-products of cognition but play a pivotal role in shaping it. They guide individuals in decision-making processes by signaling when to rely on established habits or when to seek new information [29–33].

Anger and fear, two crucial negative emotions, are often jointly triggered by different types of social threats, and in turn influence individuals’ preferences and behaviors to cope with these threats [30, 34–36]. For instance, data collected from over 12,000 respondents across six European countries demonstrated that exposure to the threat of the COVID-19 pandemic can evoke both anger and fear [28]. Another survey conducted by Erhardt et al. [34] revealed that fear and anger exert contrasting impacts on trust in the government during COVID-19 crises: fear enhances trust in the government, while anger assigns blame for adverse circumstances to the government. Building on AIT, these two specific emotional appraisals of a threat are neural correlates to different regions of the brain and leads to very different downstream behavioral consequences [37, 38]. Anger regulates behavior aimed at securing familiar goals by relying on preexisting convictions. However, encourages individuals to identify novelty and new information, fostering a learning process regarding the threat that could facilitate innovative behavioral responses [34].

Negative emotions during the COVID-19 pandemic in China

Negative emotions were widespread in times of Covid-19 [39], especially in China, where the strict "zero-COVID" prevention policy enforced for three-years. China was among the first countries to declare COVID-19 a major public health emergency in early 2020. In response, Chinese authorities swiftly implemented rigorous measures, including strict lockdowns, mass testing, extensive contact tracing, enforced quarantine for confirmed and suspected cases, and travel restrictions. These "zero-COVID" measures were maintained for three years, establishing China as a nation with one of the most enduring COVID-19 prevention strategies.

However, in December 2022, with the emergence of new variants and increased global travel, the Chinese government recognized the need for a more comprehensive approach. Subsequently, China transitioned its prevention policy from strict to lenient quickly, adapting to the changing situation [40, 41]. These policy shifts generated mixed feelings among the general population [42]. With the long-lasting restrictions, many people experienced anger and distrust in the government. They longed for a return to a more normal way of life and the resumption of previously halted activities. On the other hand, fear lingered within the public, as the liberalization of measures raised concerns about the potential for increased infections and health risks.

Moreover, conspiracy mentality is inclined to exacerbate emotional issues, particularly in the context of the COVID-19 pandemic and subsequent lockdowns. The perception of vulnerability and attack associated with conspiracy theories is linked to negative psychological states, including low self-esteem, diminished mental health, feelings of powerlessness, fear, and anger [43]. Studies also demonstrate that conspiracy mentality is a risk factor for an increased likelihood of experiencing anxiety, depressive feelings, and distress during COVID-19 lockdowns, regardless of the region [44–46]. It is evident that conspiracy beliefs not only fail to provide any beneficial consequences but also contribute to the reinforcement of negative experiences, such as anxiety, aversion to uncertainty, and existential threats [9].

Cognitive characteristics and behavioral consequences of conspiracy mentality

Previous research has identified significant associations between conspiracy mentality and various cognitive consequences, a primary one of which is distrust in institutions. Scholars have characterized conspiracy mentality as mistrust, paranoia, and defiance towards powerful groups, authority figures, and established narratives [24, 26, 47]. This distrust is reported to be linked with distrust in science in general, or susceptibility to deceptive pseudoscience [8, 48], and a reduced likelihood of adhering to government guidelines [43]. A study conducted by Chadwick et al. [49] explored the relationship between conspiracy mentality, social media use, online social endorsement, and their implication for vaccine hesitancy. The findings indicated that people scoring higher in conspiracy mentality show significantly greater intention to use social media to discourage vaccination rather than to encourage it. Nevertheless, the precise relationship among conspiracy mentality, distrust, and preventive health behaviors remains unclear, necessitating further investigation.

Conspiracy mentality is also closely associated with higher perceptions of risk or threats, as individuals often embrace conspiracy theories as a response to stress-provoking events. This inclination is driven by various social psychological motives, including the quest for subjective certainty, control, security, and the preservation of a positive self-image or group identity [13, 15]. Additionally, individuals with an underlying conspiratorial reasoning style tend to be more susceptible to strengthening their belief in conspiracy theories when confronted with external stressors [50]. Empirical evidence consistently supports the association between conspiracy beliefs and feelings of uncertainty [51], a perception of diminished socio-political control [24], heightened perceptions of risk, a sense of being under attack [5, 43, 52], and feelings of threat stemming from societal changes [53].

Aside from cognitive characteristics, conspiracy mentality might further influence individuals’ behaviors, particularly concerning preventive health behaviors. Previous research about HIV, vaccines, and other public health topics has shown that belief in conspiracy theories increases people’s resistance against important public health interventions [3, 8, 54]. Examples include resistance to get vaccinations, rejection of conventional medical treatments, and propensity for unsupported or potentially dangerous treatments based on non-mainstream belief systems [3, 6, 13, 54–56]. The influence of conspiracy mentality on medical behaviors also extends to the context of Covid-19 [57]. Studies focusing on Covid-19 related conspiracy theories have found that higher levels of coronavirus conspiracy thinking are associated with lower observance to official guidelines, reduced willingness to undergo diagnostic or antibody tests, hesitancy towards vaccination, and in some cases, intentional engagement in risky behavior [18, 43, 58].

Although multiple studies indicate that conspiracy mentality leads to fewer protective behaviors faced with diseases, its influence can vary depending on individuals’ specific beliefs. For instance, scholars have differentiated between various conspiracy claims and found that they might lead to different levels of containment behavior during a public health crisis. Individuals who perceive COVID-19 as deliberately created and distributed tend to have a more threatening perception of the pandemic and exhibit more containment behavior. On the other hand, those who consider COVID-19 a hoax perceive the pandemic as less threatening and exhibit less adherence to containment measures [59]. These findings imply that the impact of conspiracy mentality on individuals’ adoption of protective behaviors may be mediated by emotions and cognitive factors, but the underlying mechanisms need to be further explored.

Overview of the current research

While numerous studies have explored the association between conspiracy mentality and preventive health behaviors during pandemic crisis, limited research has delved into the mediating processes that explain these relationships. To date only a small handful of studies have partly investigated the mediating effects of anger, fear, institutional distrust, and perceived risk [5, 7, 10, 36]. An enlightening study by Karić and Međedović [7] examined the relationships between conspiracy beliefs, political trust and behavioral adherence. They found that reduced political trust acts as the mediating factor between conspiracy beliefs and reduced adherence to containment behaviors. Additionally, researchers have observed that distrust in authorities, feelings of powerlessness, disillusionment and perceived dangers of vaccines mediate the negative relationship between anti-vaccine conspiracy beliefs and vaccination intentions [5]. However, these studies are primarily exploratory and have not thoroughly examined the potential mediating effects of emotional and cognitive factors.

In sum, previous research indicates that conspiracy mentality affects preventive health behaviors through a complex interplay of emotional and cognitive mediating mechanisms. However, this process still requires further investigation and clarification. In the current study, we proposed an integrated emotional–cognitive mechanism to illustrate the impact of conspiracy mentality on preventive behaviors. Data was collected from Chinese citizens through an online retrospective survey during China’s "zero-COVID" policy relaxation (March 30—April 18, 2023). We therefore developed the following hypotheses in our model:

1. Hypothesis 1 (H1): Individuals scoring higher in conspiracy mentality are expected to demonstrate reduced engagement in preventive behavior during the pandemic.
2. Hypothesis 2 (H2): The impact of conspiracy mentality on individuals’ adoption of preventive behavior is proposed to be mediated by emotions, specifically anger and fear. These emotions, triggered by external threats and crises, are expected to exert their influence through cognitive factors, including perceived risk and institutional distrust, ultimately affecting individuals’ engagement in preventive behavior.
3. Hypothesis 3 (H3): The exertion process of anger differs from that of fear. Conspiracy mentality is expected to be a positive predictor of both anger and fear. Anger is anticipated to primarily lead to increased institutional distrust, and reduced protective health behavior. In contrast, fear is expected to heighten perceived risk, leading to greater adherence to preventive behavior.

The structure of the whole model is presented below (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. The proposed path model of how conspiracy mentality influences preventive behavior through mediation effects of emotions and cognitive factors.

https://doi.org/10.1371/journal.pone.0294681.g001

Study design and sampling procedure

The data used in this study was collected through an online survey of Chinese residents. The survey was conducted between March 30, 2023, and April 18, 2023, using a Chinese business survey platform (https://survey.work/survey/). Participants were recruited through convenience sampling, and the population consisted of Chinese citizens who were over 18 years old and had internet access. Informed consent was retrieved from participants before they completed the questionnaire.

A total of 2,000 questionnaires were retrieved, and after excluding invalid cases such as incomplete or low-quality replies, or replies completed in unreasonable short time, data of 1878 participants were used for further data analysis. Among the participants, 53.35% were male and 46.65% were female, which is comparable to the gender distribution in the population (51.2% vs. 48.8%). The age distribution of the participants was as follows: 43.13% were between the ages of 18 and 30, 50.47% were between the ages of 30 and 40, 5.32% were between the ages of 40 and 50, and 1.07% were above 50. In terms of education, 66.14% of the participants had a bachelor’s degree or higher, compared to 15.5% in the general population. Regarding residency, 54.74% of the participants lived in big cities, while 45.26% lived in villages or small cities, compared to 27.6% and 72.4% in the population, respectively.

The more detailed socio-demographic breakdown of the sample is presented below in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. The demographic distribution of the sample.

https://doi.org/10.1371/journal.pone.0294681.t001

Measures

Confirmatory factor analysis

To assess the validity of the measurement model, confirmatory factor analysis (CFA) was conducted before the path analysis, with 6 latent variables constructed by 31 observed indicators. Fear and anger were each represented by a single factor with no unique variables.

Convergent validity was evaluated based on the following criteria: factor loadings, reliability measured by Cronbach’s α, and average variance extracted (AVE). Specifically, two items from the conspiracy mentality scale, two items from the institutional distrust scale, and one item from the preventive behavior scale were excluded from the model since their factor loadings were lower than 0.6. All the factor loadings of the remaining observed indicators were higher than 0.6. All AVE values for the measurements exceeded 0.5. The Cronbach’s α values of all the measurements exceeded 0.8 as displayed above, indicating a good convergent validity. Discriminant validity was evaluated by contrasting the values of AVE with the maximum shared squared variance (MSV). These comparisons revealed that the AVE values for all the measurement variables were greater than their corresponding MSV values, indicating good discriminant validity.

With regards to the overall model fit, results of indexes proved to be satisfactory. Chi-square test: CMIN/DF = 5.476, RMSEA = 0.049 (<0.05), GFI = 0.924 (>0.9), AGFI = 0.908 (>0.9), CFI = 0.962 (>0.95), SRMR = 0.051 (<0.06). We therefore concluded the measurement model is fit for further analysis.

Path analysis

We constructed the whole structural equation model to test our hypotheses and examine the mediating processes after the validation of the measurement model. Correlational analyses showed significant associations between participant demographics, including age and educational level, with conspiracy mentality, fear and anger. Therefore, age and educational level were incorporated in the path analysis model as control variables. Educational level was quantified in terms of years of education, reflecting the highest level of educational attainment.

The fit indices of the model are as follows. Due to the sensitivity of χ² to sample size, indexes including RMSEA, GFI, AGFI, CFI, and SRMR were also reported. According to Kline [63], if the RMSEA value is lower than 0.05, it can be accepted as a good approximate fit, and the range between 0.05 and 0.08 suggested a reasonable fit. Hu and Bentler [64] noted that a cut of value close to 0.95 for CFI and a cutoff value close to 0.08 for SRMR can be considered as indicators for a satisfactory model fit. According to these criteria, the path analysis proved to be an overall good fit: CMIN/DF = 4.66 (<5), RMSEA = 0.044 (<0.05), GFI = 0.930 (>0.9), AGFI = 0.916 (>0.9), CFI = 0.964 (>0,95), and SRMR = 0.036 (<0.05).

The control variables included in the model were also examined with their relationships with measurement variables. Results showed that educational level had a negative correlation with both anger (β = -0.105, se = 0.020, p < 0.001) and institutional distrust (β = -0.074, se = 0.018, p < 0.001). Age had a positive correlation with perceived risk (β = 0.094, se = 0.022, p < 0.001). Results indicated that people with higher educational attainment were associated with fewer emotions of anger and less institutional distrust, while elderly people experienced higher perceptions of risk faced with the pandemic.

H1 posits that individuals scoring higher in conspiracy mentality are expected to demonstrate reduced engagement in preventive behavior during the pandemic. In support of H1, the total effect of conspiracy mentality on preventive behavior was -0.107 (se = 0.026, p < 0.001, 95% bias-corrected confidence interval = -0.157, -0.055), signifying that conspiracy mentality is associated with significantly less engagement in preventive behavior.

H2 probes into the mediating effects of emotions and cognitive factors in the process by which conspiracy mentality leads to reduced adoption of preventive behavior. To validate the mediation effects, a bootstrap test was conducted using 2,000 bootstrap resamples. Results showed significant indirect effects from conspiracy mentality to preventive behavior (β = -0.110, se = 0.024, p < 0.001, 95% bias-corrected confidence interval = -0.158, -0.062). These findings indicated that the chains of anger leading to institutional distrust and fear leading to perceived risk serve as significant mediators between conspiracy mentality and engagement in preventive behavior, thus supporting H2. Accounting for these significant mediational pathways, the direct effect from conspiracy mentality to preventive behavior was not significant (β = 0.002, se = 0.032, p = 0.898, 95% bias-corrected confidence interval = -0.056, 0.067). This suggests the effect from conspiracy mentality to preventive behavior was fully mediated by the two pathways. Additionally, it is noteworthy that the effect from conspiracy mentality to media trust was partially mediated by anger, with a significant yet relatively small indirect effect of 0.097 (the ratio of indirect effect to the total effect is 16.1%). On the other hand, fear fully mediated the relationship between conspiracy mentality and perceived risk (direct effect = 0.059, se = 0.027, p = 0.021, 95% bias-corrected confidence interval = 0.009, 0.115, which was not significant accounting for the mediation effect). This implies that the effect of conspiracy mentality on perceived risk primarily operates through the path of fear, while increased institutional trust is more directly influenced by conspiracy mentality itself, with only partial mediation by the emotion of anger. The details of the total, direct and indirect effects of the proposed mediational pathways are presented in Table 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Total, direct, and indirect effects of conspiracy mentality and mediators.

https://doi.org/10.1371/journal.pone.0294681.t003

H3 explores the divergent effects of anger and fear in the mediational processes. It is hypothesized that conspiracy mentality is simultaneously associated with increased anger and fear. While anger predicts increased institutional distrust, and reduced protective health behavior, fear is expected to be linked with heighten perceived risk, leading to greater adherence to preventive behavior. The results of path analysis supported the hypotheses. Conspiracy mentality was found to be a significant positive predictor for both anger (β = 0.420, se = 0.023, p < 0.001) and fear (β = 0.393, se = 0.024, p < 0.001). Anger was positively correlated with institutional distrust (β = 0.231, se = 0.024, p < 0.001), and institutional distrust was a negative predictor of preventive behavior (β = -0.303, se = 0.032, p < 0.001). On the other hand, fear was found to be a positive predictor of perceived risk (β = 0.425, se = 0.026, p < 0.001), and perceived risk was associated with higher engagement in preventive behavior (β = 0.343, se = 0.030, p < 0.001). To conclude, higher levels of anger among people with elevated conspiracy mentality sparked more distrust in mainstream accounts and official media, ultimately resulting in reduced engagement preventive behavior. Conversely, the heightened fear, driven by conspiracy mentality during times of threat, led to greater perceptions of risk and increased participation in preventive behavior.

Fig 2 presents the revised model with detailed validating results of the hypotheses.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Path model estimation results.

Significance: *p < 0.05, **p < 0.01, ***p < 0.001. H1 and H2 tested total and mediation effects, therefore were not directly reflected in the path model. The detailed results of H1 and H2 can be found in Table 3.

https://doi.org/10.1371/journal.pone.0294681.g002