This study examined the relationships between paranormal beliefs and gender and thinking styles. London-based university students majoring in psychology (N = 158) filled in two questionnaires: the Rational-Experiential Inventory 40 and the Revised Paranormal Belief Scale. The results showed that intuitive thinking was positively correlated to paranormal beliefs, and rational thinking was negatively correlated to paranormal beliefs. Women’s higher number of paranormal beliefs was partially explained by their greater intuitiveness compared with men. Due to the weakness of the correlations found, the discussion section focused on the potential shortcomings of the two measures used, the lack of consensus in the literature on thinking styles, and the importance of taking cultural aspects into account when measuring paranormal beliefs and thinking styles.
Keywords: Paranormal beliefs, thinking styles, rationality, critical thinking, intuitive thinking, dual-process model, gender, cultural aspects.
For over four decades (Salter & Routledge, 1971), the debate about the relationship between thinking style and paranormal beliefs has raged in the scientific community. It could be argued that one of the factors sparking this debate is that belief in supernatural abilities, paranormal phenomena, superstitions, religious entities and extra terrestrial beings can be found in almost all cultures worldwide. According to Zuckerman (2007), roughly 90 per cent of the world’s population hold religious beliefs or believe in some kind of supernatural being. As paranormal, superstitious, supernatural and religious beliefs are neither falsifiable, according to Popper (1963), nor able to stand the test of proper scientific scrutiny, as per the challenge posed by Randi (2013), one could infer that they are vulnerable to scientific disconfirmation and rejection based on critical thinking. Since the common denominator for these beliefs is a lack of scientific evidence to support them, the supernatural, superstitious, religious, theistic, paranormal and magical could all be considered to be part of a broader spectrum of irrational beliefs.
Some of the more recent studies, including those of Aarnio and Lindeman (2005; 2006; 2007), have used self-reported data by utilising the Rational–Experiential Inventory 40 and the Revised Paranormal Belief Scale, and have provided evidence for paranormal beliefs being tied to the participants’ style of thinking. The Revised Paranormal Belief Scale (RPBS), based on the original Paranormal Belief Scale (Tobacyk & Milford) and developed by Tobacyk (2004), has been employed by a significant number of researchers studying paranormal belief and has been backed by a lot of evidence supporting the validity of the scale. The definition of paranormal phenomena in this context, as well as in the context of the current study, is ‘phenomena that, if genuine, would violate the basic limiting principles of science’ (Broad, 1953). The Rational-Experiential Inventory (REI-40) developed by Pacini and Epstein (1999) has also been widely used to measure thinking styles, and builds on the Faith in Intuition (FI) scale (Epstein, Pacini, Denes-Raj & Heier, 1996; Pacini & Epstein, 1999) and the Need For Cognition (NFC) scale (Cacioppo & Petty, 1982).
The current study will begin by examining the current literature on thinking styles and paranormal beliefs and the evidence from using the RPBS and REI-40 as scales of measurement. The aim of the study is to support the current evidence by testing the scales on a less homogenous participant sample than those employed by previous studies in order to determine the validity of the scales and previous studies. As an additional test for validation, the current study will also look at gender differences in thinking style and paranormal beliefs.
Since the introduction of the dual-process model of rational processing (Sun, 2002), much research has been conducted on decision-making, risk assessment and problem solving in the past decades and has produced theories on two types or systems of reasoning (Epstein, 1994; Sloman, 1996; Kahneman, 2003; Evans, 2008; Stanovich, 2009). The first type, which is based on fast, intuitive, heuristic, associative and unconscious processing, is labelled System 1 processing. System 2 processing involves a broader assessment of the situation and requires a more thorough and time-consuming analysis than System 1 processing and a critical evaluation of initial intuitions. The labels for System 1 and System 2 were coined by Stanovich and West (2000) to clear up the confusion caused by the broad variety of descriptions used by researchers in the field, such as ‘intuitive’ and ‘experiential’ (System 1), and ‘rational’ and ‘analytic’ (System 2).
With reference to gender differences in thinking style, studies carried out by Pacini and Epstein (1999), as well as Lieberman (2000) found that women tend to think more intuitively (System 1) than men, whereas men have a tendency towards more analytical (System 2) thinking than women.
Since the evidence indicates that System 2 thinking is more useful for making rational decisions, why do people often seem to use System 1 thinking? Several researchers have found that the initial intuition used in System 1 processing typically involves accessible beliefs and knowledge. Studies in support of this finding report that participants who opt for System 1 processing do so because the choice ‘seems right’ or is perceived to be self-evidently valid. The researchers found that this intuitive way of processing information requires less cognitive effort (Epstein et al., 1996; Stanovich & West, 2000; De Neys, 2006; Morewedge & Kahneman, 2010; Thompson, Prowse, Turner & Pennycook, 2011).
What is the relationship, then, between the dual-process model and paranormal beliefs? Epstein et al. (1996) found a positive correlation between System 1 thinking and paranormal beliefs. These findings were supported by further evidence in later studies (Gianotti, Mohr, Pizzagalli, Lehmann & Brugger, 2001; Aarnio & Lindeman, 2007). They reported that the rationale for these studies was that the justification for paranormal beliefs is based on subjective experience without the requirement of objective evidence to support the belief. Given that there is no solid evidence to support paranormal beliefs, what makes them attractive to people? Some studies have described that the attraction of supernatural beliefs is due to people finding them minimally counterintuitive (Barrett, 2000; Boyer, 2001; Norenzayan, Atran, Faulkner & Schaller, 2006; Pyysiäinen & Anttonen, 2002). However, this is in stark contrast to research findings revealing that an increasing number of people find many supernatural concepts directly counterintuitive (Hunsberger & Altemeyer, 2006). Whether paranormal beliefs are minimally counterintuitive or directly so, what would be the reason for people to retain their paranormal beliefs in favour of a rational or even a more intuitive belief? Beins (2002) points out that the reason for many people retaining their paranormal beliefs despite being faced with contradictory evidence could be that their beliefs reduce uncertainty.
Having looked at the positive correlation between System 1 thinking and paranormal beliefs, it would be interesting to solidify the relationship between thinking style and paranormal beliefs by examining the corollary hypothesis. Indeed, evidence produced by Caplovitz and Sherrow (1977) showed a negative correlation between intellectual ability and supernatural beliefs among university students. In the light of this evidence, Pennycook, Cheyne, Seli, Koehler and Fugelsang (2012) found that individuals who were more willing to engage in System 2 processing were less likely to hold paranormal beliefs. The reverse also holds true, as several studies have found that nonbelievers engage in more System 2 thinking in general (Hunsberger & Brown, 2001; Beit-Hallahmi, 2006; Caldwell-Harris et al., 2010). Hunsberger and Altemeyer (2006) found that the reasons for the rejection of these beliefs have are be scientific and rational. Based on Stanovich and West’s (2008) findings that logical inference is crucial when testing for intelligence, several studies have found a negative relationship between supernatural beliefs and intellectual ability (Bertsch & Pesta, 2009; Lynn, Harvey & Nyborg, 2009; Reeve, 2009; Lewis, Ritchie & Bates, 2011). It is possible that there is a negative correlation between supernatural beliefs and reasoning tasks related to System 2 thinking. Indeed, this hypothesis is supported by evidence reported by Shenhav, Rand and Greene (2011).
In light of the evidence indicating a negative relationship between System 2 thinking and paranormal beliefs, it is important to determine what exactly constitutes paranormal beliefs. The broad spectrum of these beliefs include, but is in no way limited to topics such as astrology, dowsing, clairvoyance, ESP, extra terrestrial beings, ghosts, voodoo, witchcraft, telekinesis, certain conspiracy theories, and the more commonly accepted traditional theistic beliefs, such as the Abrahamic religions. These religious belief systems are typically tied to a range of other specific supernatural beliefs such as miracles, the afterlife, demonic possession, exorcism, blessings and curses. Moreover, various rituals and practices, such as prayers, baptisms and weddings, often accompany these theistic beliefs. The fact that some of these practices have become legal institutions in Western countries illustrates the extent to which they have been accepted into the public sphere.
Although the reasons for irrational beliefs being attractive to people have been covered, the question of how they came about still remains unanswered. Recent studies on the origins of these irrational beliefs have indicated that they are an evolutionary by-product of the way people process information, which tends to rely heavily on the System 1 thinking style to which paranormal beliefs are positively correlated (Guthrie, 1993; Boyer, 1994; Barrett, 2000; Lawson, 2000; Pyysiäinen, 2001, Atran, 2002; Evans, 2003; Frey, 2009). Evans (2008) suggested that System 1 processing naturally precedes System 2 processing due to the fact that humans are cognitive misers, which could be interpreted as people having a natural inclination towards paranormal beliefs. However, Mithen (2002) argued that the success of the human species lends credence to the adaption of System 2 thinking, which provided the species with higher cognitive abilities than other hominids.
It would be interesting, then, to examine how the further development of these cognitive abilities through education, could impact on irrational beliefs. Sampling participants from various educational backgrounds, ranging from university students to trade school students, Za’rour (1972), as well as Otis and Alcock (1982), suggested that the critical thinking skills taught to college and university students could have an impact on their relationship with paranormal beliefs, as these skills would better equip them to apply System 2 processing to irrational claims. This argument is supported by Gray and Mill (1990), as well as Musch and Ehrenberg (2002), whose findings have provided evidence for a positive relationship between paranormal beliefs and a lack of critical thinking skills. These findings contradict earlier studies that reported the opposite, and could indicate that the current evidence in the support of the hypothesis is insufficient (Parida, 1962; Jahoda, 1968; Pasachoff et al., 1970; Salter & Routledge, 1971; Wolfradt & Vyse, 1997; Roe, 1999; Oubaid, Straube, Bischoff & Mischo, 1999). However, Aarnio and Lindeman (2005) argue that this could be due to methodological issues in the earlier studies. The first reason they give is that much of the data provided could be out of date as the educational environment was very different in the 1970s. The second reason given is that there is a lack of detail regarding the methods and results of these studies compared with more recent studies and makes the evidence difficult to compare. Finally, the studies included relatively small numbers of participants and the contradictory findings could be due to the sample size. It could thus be argued that the reason for the negative relationship between paranormal beliefs and level of education has not yet been fully determined, but an increasing number of later studies point to a negative correlation between level of education and irrational beliefs.
Two of these studies were carried out by Beit-Hallahmi (2006) and Zuckerman (2007). They found that in modern societies, such as the industrialized Western countries − among the most progressive in terms of education, the number of people who find religious beliefs directly counter to their intuition is on the rise. This was illustrated by a census carried out in 2011 by The Office for National Statistics, which showed that the number of atheists and agnostics was growing in the UK. However, contradicting evidence is to be found in both earlier and later studies: the number of people holding superstitious and paranormal beliefs is also on the rise: recent figures show that as many as 77 per cent of British residents reported that they are at least slightly superstitious (such as afraid of walking under ladders) (Wiseman, 2003) and 52 per cent reported that they firmly believe in supernatural elements, like communicating with the dead (ASSAP, 2013). According to Dave Wood, the chairman of the ASSAP (Association for the Scientific Study of Anomalous Phenomena), these figures are significantly higher than the historical average. In North America, the figures for people with paranormal beliefs are similar to those in the UK: Orenstein (2002) and Rice (2003) found that over 50 per cent of adults hold supernatural beliefs such as telepathy and clairvoyance. Moore (2005), referencing a survey conducted by Gallup, found that as many as 73 per cent of US citizens hold at least one paranormal belief, such as belief in ghosts or psychic powers.
Regarding gender differences in paranormal belief, results of several studies on the subject have shown that women tend to hold more beliefs than their male counterparts, with the exception of beliefs in UFOs and extra terrestrial life forms (Tobacyk & Pirttilä-Backman, 1992; Vyse, 1997; Rice, 2003; Blackmore, 2012). This might be explained by the gender differences in thinking style, mentioned earlier.
It could be argued that the reasons for these seemingly contradictory findings in people’s paranormal beliefs involve numerous influential variables such as affective, experiential, family, institutional, developmental, and cultural factors. Although this leaves the connection between intuitive thinking and paranormal beliefs open to interpretation, Sutherland (1992) reports that the many academic works produced on decision-making, including that of Tversky and Kahneman, clearly indicate that human intuition has a considerable effect on reasoning and rational processing. In light of this, the current study aims to support the previous findings by testing whether a critical thinking style is negatively correlated to paranormal beliefs, whether education has an impact on paranormal beliefs, and whether there is a gender difference in paranormal beliefs. To summarise, the following hypotheses are addressed:
The participants (N = 158) were 18–50-year-old (M = 22.9 years, SD = 6.6 years) full-time university students majoring in psychology, of whom 80.4 per cent were women. The participants were sampled from an undergraduate body of students consisting of the following ethnic breakdown (self-report): Asian: 32 per cent, Black: 27 per cent, Mixed: 12 per cent and White: 29 per cent.
The participants were recruited at the University of East London. They were asked to participate immediately before a lecture in critical thinking when the questionnaires were handed out along with the brief and consent form.
The participants were briefed prior to being handed the questionnaires, and were asked to read and fill out an ethics form before proceeding with the questionnaires.
Measures and Materials
Paranormal beliefs were measured with the Revised Paranormal Belief Scale (Tobacyk & Milford, 1983; Tobacyk (1988). The items of the RPBS are divided into seven subscales consisting of traditional religious beliefs, spiritualism, extraordinary life forms, psi, witchcraft, precognition, and superstition. The RPBS consists of 26 items, and is measured by means of a seven-point scale (1 = strongly disagree, 7 = strongly agree). Examples of items measuring paranormal beliefs are ‘Witches do exist’ and ‘There is a heaven and a hell.’ Rational and experiential thinking styles were measured with the Rational-Experiential Inventory 40 (Pacini & Epstein, 1999). The REI-40 consists of 40 items (20 for rational and 20 for experiential), and is measured with a five-point scale (1 = strongly disagree, 5 = strongly agree). Example items measuring rational and experiential thinking are ‘I think it is foolish to make important decisions based on feelings’ and ‘I tend to use my heart as a guide for my actions’.
The scores obtained from the RPBS and REI-40 questionnaires were entered into SPSS, where the appropriate scores were reversed as per the questionnaire descriptions. Table 1 shows the correlations between REI-40 scores and RPBS scores, as well as the correlations between the four subsections of the REI-40.
Regarding the relationship between the scores derived from the REI-40 and the RPBS, the results showed that there was no correlation between the two variables Rational Engagement and RPBS Total (r = 0.061, n = 146, p = 0.468), but the results were not significant. There was a weak negative correlation between the two variables Rational Ability and RPBS Total (r = -0.248, n = 146, p = 0.002) with highly significant results. There was a weak negative correlation between the two variables Rational Total and RPBS Total (r = -0.217, n = 136, p = 0.011) and the results were significant. There was no correlation between the two variables Experiential Engagement and RPBS Total (r = 0.083, n = 147, p = 0.317), but the results were not significant. There was a weak positive correlation between the two variables Experiential Ability and RPBS Total (r = 0.109, n = 155, p = 0.177), but the results were not significant. There was a weak positive correlation between the two variables Experiential Total and RPBS Total (r = 0.124, n = 147, p = 0.133), but the results were not significant.