In the literature, illusory beliefs are also defined as beliefs in the paranormal [5, 6]. Phenomena that can be included in the realm of the paranormal are very diverse, and the definition of paranormal belief itself has been revised many times, with regard primarily to the following points: 1. that which cannot be explained in terms of contemporary science; 2. that which can be explained only by significant revisions of limiting and fundamental scientific principles; 3. incompatibility with normative perceptions, beliefs, and expectations about reality [7, 8, 5]. Therefore, it is a complex construct and one that is difficult to operationalize; however, for many years already, there has been a scholarly consensus that it is a multidimensional construct. Tobacyk [8, 9] has individuated the following dimensions: Traditional Religious Belief, Psi, Witchcraft, Superstition, Spiritualism, Extraordinary Life Forms, and Precognition. The relationship between paranormal beliefs and health has been investigated in several studies, which have found the existence of a significant and positive correlation with mental illness [10] and with manic-depressive experiences [11]. However, contradictory results have been found with respect to the relationship to neuroticism [12, 13] and anxiety [14, 15]. Other studies have found that some types of beliefs (eg., religious and fatalistic beliefs) may inhibit the utilization of healthcare and healthcare-related behaviors [16, 17]. More recently, illusory beliefs have been considered in the literature, specifically in relation to the sphere of health [4, 18], such as the beliefs and practices of timely healing, disease prevention, and the general promotion of health. In the case of paranormal beliefs related to health, several other dimensions have already emerged from more general studies of beliefs in the paranormal, such as Religious Beliefs, Superstitious Beliefs, Extraordinary Events Beliefs, and Parapsychological Beliefs. However, another dimension also appears: that of Pseudo-Scientific Beliefs of a Bio-Medical Nature, which refer to the negative impact on human health of specific categories considered socially deviant or marginal (homosexual, immigrant) and to the threats to health posed by hereditary transmission or genetic contamination (interracial relationships). In these studies, it has been shown that these specific beliefs are positively associated with general illusory beliefs, thereby not only those related to health; furthermore, they are also positively associated with an external health-related locus of control, and negatively associated with the internal health-related locus of control. With respect to sex-based differences, men, both adolescents and adults, have more confidence in medicine as a science and have a tendency to rely on a bio-medical approach in relation to the protection of the species, while adult women have higher levels of religious beliefs [4, 18].

From a social-cognitive perspective, the function of illusory beliefs is fundamentally adaptive; in fact, Taylor and Brown [19], in speaking about the relationship between health and the paranormal, suggest the idea of the self-serving illusion, emphasizing that illusory beliefs, even as they are definitively false beliefs, serve an important function for mental health by creating a “filter” through which reality acquires order and meaning. Grimmer and White [20] have also demonstrated the relevance of such belief systems for adaptation in the realm of health and illness, with specific reference to various dimensions.

From a psycho-dynamic point of view, instead, Irwin [21] has formulated the psychodynamic functions hypothesis, which sees belief in the paranormal on par with a coping strategy, like that of avoidance, which occurs in the treatment of post-traumatic stress.

Considering the role that illusory beliefs can have in the behavioral choices of individuals, the first objective of the present study is to investigate the relationship between Pseudo-Scientific Beliefs of a Bio-Medical Nature and Social Dominance Orientation, as well as with the God-Centered Health Locus of Control, Coping by Turning to Religion, and Mental Illness. We have hypothesized the existence of positive interrelationships among Pseudo-Scientific Beliefs of a Bio-Medical Nature and God-Centered Health Locus of Control, Social Dominance Orientation, Coping by Turning to Religion, and Illness.

Furthermore, in the perspective of developing pscyho-educational intervention programs that could raise awareness of the socio-cultural aspect of these beliefs, the second goal is understanding the role of personality variables (Social Dominance and Locus of Control) with regard to Pseudo-Scientific Beliefs and their function, as well as turning to religion coping in relation to perceived Illness. More specifically, as far as the determining factors of illness, which are considered an outcome variable, we have hypothesized a model (Fig. 1) in which: Social Dominance and God-Centered Health Locus of Control have been inserted as antecedents of Pseudo-Scientific Beliefs of a Bio-Medical Nature; God-Centered Health Locus of Control and Pseudo-Scientific Beliefs are considered predictors of Coping by Turning to Religion; and Pseudo-Scientific Beliefs and Coping by Turning to Religion are predictors of Illness. We hypothesize, furthermore, that the relation among the variables as relates to the religious realm could be different for religious believers (Catholics) and atheists.

Fig. (1). Conceptual Model of the Structural Equation Model.

A convenience sample of people hailing from Southern Italy was obtained, composed of 451 people. 19 subjects were eliminated, as they declared belief in a religion other than Catholicism. The 432 participants in the study were mostly female (76.8% female; 23.2% male), with ages ranging between 18 and 70 years (M = 24.9, SD = 11.6). 66.2% of the participants were Catholic and 33.8% were atheist. 57.0% of the participants have a center-left political orientation, 28.9% center, and 14.1% have a center-right political orientation.

Participation in the study was voluntary and anonymous, and participants were encouraged to answer as truthfully as possible. Furthermore, all the study participants provided written informed consent.

Making use of a quantitative methodology, a questionnaire, including validated instruments in the Italian language, was specifically designed for the study.

Paranormal Health Beliefs Scale [4, 18]. For the current study, we used the following sub-scale for Pseudo-Scientific Beliefs of a Bio-Medical Nature, which refer to beliefs regarding the negative contribution to the health of the human species of specific categories that are considered to be deviant or marginal social groups, and to health threats deriving from hereditary transmission or genetic contamination (ex., “Persons affected by generational evil constitute a threat for the health of the human species” and “Relationships between people of different races may be harmful to health”; α = 0.65). Each sub-scale consists of 4 items rated on a 5-point Likert scale (from 1 = “Strongly Disagree” to 5 = “Strongly Agree”): high scores relate to high levels of illusory beliefs referring to health.

Social Dominance Orientation Scale [46, 48], assesses the preference for inequality among social groups and the superiority of some people over others, with 16 items such as “Some groups of people are simply inferior to other groups,” and “No one group should dominate in society.” All items were measured on a 7-point Likert scale, anchored at “Strongly Agree” and “Strongly Disagree.” Higher scores indicate higher levels of SDO. The scale had an internal consistency coefficient of 0.73.

Health Locus of Control Scale [42, 43]. For the current study, we used the following sub-scale of the God-Centered Health Locus of Control scale for the measurement of the religious orientation of locus of control as related to health (ex., “If God wants, my physical health can improve” and “My physical health can be safeguarded with God's help”; α = 0.89). Each sub-scale consists of 2 items rated on a 5-point Likert scale (from 1 = “Strongly Disagree” to 5 = “Strongly Agree”). Higher scores indicate higher levels of Locus of Control.

Coping Orientation to Problems Experienced [28, 52]. For the current study, we used the following sub-scale of the Turning to Religion scale (ex., “How often do I try to find comfort in my religion?”, and “How often do I look for help in God?”; α = 0.96). Each sub-scale consists of 4 items rated on a 6-point Likert scale (from 1 = “I never do that” to 6 = “I always do that”). Higher scores indicate higher levels of coping strategies.

General Health Questionnaire [53, 54] is aimed at detecting nonpsychotic psychiatric morbidity in the general population (ex., “Have you recently been able to concentrate on whatever you’re doing?”, and “Have you recently been able to enjoy your normal day to day activities?”). The scale consists of 12 items rated on a 4-point Likert scale rating (from 1 = “Not at all” to 4 = “Much more than usual”), distinct in: Somatic Symptoms and Social Dysfunction. For the purposes of this study, the overall scale score was considered. Higher scores indicate greater levels of general psychiatric distress. Internal consistency reliabilities have been found to be 0.86.

Finally, the questionnaire included a section for the collection of socio-demographic data (gender, age, religion, and political orientation). Each questionnaire took approximately 30 minutes to complete.

The study protocol was approved by the Local Ethics Committee, and the study was conducted according to APA ethical standards. The study conformed to the ethical principles of the 1995 Helsinki Declaration.

In order to calculate the reliability of the scales, we used Cronbach’s alpha. An internal consistency greater than .70 is thought to be necessary for the psychological scale, even if an alpha between 0.60 to 0.69 would be considered acceptable [55].

We used descriptive statistics to analyze the characteristics of the respondents and the study variables. Pearson’s correlation coefficient was used to determine the relationship between all the variables (p-value < 0.05). Finally, Structural Equation Modeling (SEM) was used to test structural relationships. Goodness-of-fit of the model was assessed as indicated by a chi-squared distribution and the degrees of freedom (χ²/df ≤ 3), a Standardized Root Mean Square Residual (SRMR ≤ 0.08), Comparative Fit Index (CFI > 0.90), and Non-Normed Fit Index (NNFI > 0.90). Results of the Root Mean Square Error of Approximation (RMSEA) are considered to be good if they are ≤ 0.05 and reasonable if they are ≤ 0.08. Because the different indices assess different aspects of goodness-of-fit, evaluating multiple fit indices simultaneously is recommended [56-58]. Satisfactory models should show consistently good-fitting results on many different indices.

Survey data were then entered into SPSS 18.0 [59] and Lisrel 8.54 [60] databases and checked/verified by project staff for accuracy.

Means, Standard Deviations, and correlations are shown in Table (1). The results for zero-order correlations between the instruments indicate that Pseudo-Scientific Beliefs are not correlated with Illness, but are highly positively correlated with Social Dominance Orientation (0.31**), Coping by Turning to Religion (0.23**), and God-Centered Health Locus of Control (0.16**). Indeed, Illness was highly positively correlated only with Social Dominance Orientation (0.10**) and Coping by Turning to Religion (0.10**).

We used structural equation modeling to test the structural relationships. The hypothesized model for predicting illness was tested Fig. (2) and the results confirmed our model, with good fit between the theoretical and the empirical models: χ²(df) = 184.30(71), p = n.s.; χ²/df = 2.59; CFI = 0.94; NNFI = 0.92; RMSEA = 0.06 [0.05, 0.07]; SRMR = 0.07; GFI = 0.94; AGFI = 0.91. As hypothesized, Social Dominance is a strong predictor of Paranormal Health Beliefs along with External Locus of Control, which is a very strong predictor for coping by turning to religion. In turn, Coping by Turning to Religion is also predicted by Pseudo-Scientific Beliefs which, however, do not predict Illness, upon which only Coping by Turning to Religion has bearing.

Fig. (2). Model for the Prediction of Illness. Note: *p < .05; **p < .01.

Furthermore, the model was tested separately on Catholics and atheists. Applying the model only to Catholics (N= 286), the relations between the variables are confirmed, and the model as a whole shows good indices of fit:: χ²(df) = 124.12(71), p = n.s.; χ²/df = 1.75; CFI = 0.95; NNFI = 0.93; RMSEA = 0.05 [0.04, 0.07]; SRMR = 0.06; GFI = 0.94; AGFI = 0.91. Applying the model only to atheists (N= 146), the indices of fit are not satisfactory: χ²(df) = 138.42(71), p = n.s.; χ²/df = 1.94; CFI = 0.82; NNFI = 0.78; RMSEA = 0.08 [0.06, 0.10]; SRMR = 0.11; GFI = 0.88; AGFI = 0.82.

This study has several limitations. First, we must consider the relatively small number of participants and the limited territorial origin of the participants, as well as the lack of diversity in terms of demographic characteristics (sex and political orientation). Further limits include the cross-sectional characteristics of study and self-report data.

In the future, a longitudinal study should be envisaged with the involvement of a greater number of participants, balanced in terms of gender and political orientation, and coming from different territorial contexts. Furthermore, future investigations could integrate quantitative with qualitative data.

In the future, further studies would be advisable if they could verify the hypothesis of the mediational role of paranormal health beliefs in the relation between social dominance and illness. It would also be useful to deepen understanding of the relationships between pseudo-scientific beliefs of a bio-medical nature and other beliefs, such as belief in genetic determinism. We would also advise studies on the predictive role of pseudo-scientific beliefs of a medical nature regarding support for, or opposition to, welfare policies, and studies on the relationship between social dominance and well-being, taking into account the sub-dimensions of well-being (especially social well-being).

Finally, it would be interesting to test this model on groups who have been differentiated by types of religion and participation in religious activities, along with a further understanding of the existence of these relationships in specific social categories such as teachers [66, 67] and doctors [68], for the effects that their ill health may have on students and patients.