This post is in response to Religion and Reason by R. Douglas Fields

Simply viewing this sculpture in an experiment reduced religious belief.

Recent papers (Gervais & Norenzayan, 2012; Pennycook, Cheyne, Seli, Koehler, & Fugelsang, 2012; Shenhav, Rand, & Greene, 2011) have suggested that religious and paranormal beliefs are supported by “intuitive” thought processes and that engaging in “analytical” thought processes can weaken these beliefs, at least temporarily. These studies have fascinating implications for the development of non-rational beliefs and suggest possibilities for challenging such beliefs. These studies consider intuition and analytical thought as opposed systems, and while this is frequently applicable, previous research has found that in some people these two modes of thought co-exist to a high degree and are associated with supernatural beliefs.

These studies are based on a “dual process” model of thought that proposes the existence of two parallel information processing modes. The first mode, sometimes called the experiential system, is intuitive, emotional and provides immediate responses. The second mode, sometimes called the rational system, is reflective and logical, and involves deliberation, and hence takes more time to process information than the intuitive system. There are individual differences in the degree to which people prefer to use either of these systems. Studies have found that people who show a greater inclination towards the use of intuitive thinking were more likely to endorse belief in God and the afterlife (Shenhav, et al., 2011) as well as in paranormal phenomena, including psychic powers, witchcraft, astrology and so on (Pennycook, et al., 2012). Pennycook et al. even classified belief in God along a spectrum from most conventional (a personal God exists), through the less conventional (belief in an impersonal higher power or an inactive God), and through into degrees of disbelief (agnosticism and atheism). They found that more conventional God beliefs were associated with more intuitive responses, whereas less conventional beliefs and disbelief were associated with more analytical responses.

Even more intriguing were the results of experimental interventions to manipulate intuitive versus analytical thinking. Belief in God was strengthened by a writing intervention designed to either increase confidence in one’s intuition or reduce confidence in one’s rationality (Shenhav, et al., 2011). Conversely, belief in God was weakened by decreasing confidence in intuition or increasing confidence in rationality. A series of experiments in which rationality was primed were also found to decrease belief in God (Gervais & Norenzayan, 2012). The primes used included viewing a picture of Rodin’s “Thinker”, reading words associated with analytical thinking (think, reason, analyze, ponder, rational), and, remarkably, reading a questionnaire in a difficult to read font. Previous research has found that information presented in a difficult to read font improves subsequent performance on tests of analytical thinking. Notably, none of these interventions involved any form of argumentation against religious belief or any mention of belief at all. People were simply asked to perform these tasks and then they answered questions about their beliefs.  

These findings have interesting implications for attempts at persuasion for or against religious beliefs. Encouraging people to trust their intuitions or gut feelings in matters unrelated to religion could increase a person’s subsequent receptivity to religious or supernatural ideas. Conversely, encouraging intellectual discussion or even having people read unrelated material in a hard-to-read font could elicit scepticism about such things. The studies cited could not address how durable these effects are likely to be, so it is possible that such interventions might have only transient effects.


The quote is attributed to Martin Luther, 1569.

The findings of the studies do provide evidence that religious and supernatural beliefs are associated with intuitive thinking and that scepticism about religion is associated with analytical thinking. However, it would be premature to conclude that such beliefs are best viewed in terms of a dichotomy in which intuitive and analytical thinking are by nature opposed to each other. Such a dichotomy seems to be implied by statements by Pennycook et al. that: “initial intuitions during problem-solving often pre-empt further analysis” (p. 336) and “An analytic cognitive style will typically involve a broader assessment of problem elements as well as an examination and critical evaluation of intuitions.” The task used to assess preference for intuitive or analytical thinking actually involves asking people to solve problems where there is an intuitively appealing but incorrect answer, so that to reach the correct answer the person must reject their initial intuition. The nature of the task itself therefore sets up a dichotomy in which intuition and reflection are incompatible. This may well apply much of the time but there is intriguing evidence that the two modes of thought sometimes operate in a parallel rather than a conflicting manner. This parallel operation has been found to have implications of its own for supernatural beliefs.

The dual-process model originally proposed that the two modes are independent of each other and as a result although some people habitually prefer one mode over the other, some people actually prefer to use both, while some people have little preference for either. One study sorted people into four clusters based on their respective preferences for the two modes of thought and examined their patterns of belief (Wolfradt, Oubaid, Straube, Bischoff, & Mischo, 1999). The cluster who had high preferences for both intuitive and analytical thought (“complementary thinkers”) actually had the highest rates of paranormal belief of all four clusters and also scored higher on measures of magical thinking compared to the cluster with high analytical and low intuitive thinking (“rational thinkers”). Wolfradt et al. suggested that: “A simultaneous processing style leads potentially to a disinhibition of associations which fosters irrational thinking” (p. 828).

The potentially complementary relationship between intuitive and analytical modes of thought seems to shed some light on religious beliefs among scientists. Scientists are vocationally committed to empiricism and therefore would be expected to be naturally high on analytical thinking. Scientists are generally expected to reject beliefs that are not based on evidence and surveys have found that the majority of scientists are not religious. However, there are a substantial minority of scientists who are religious (one study found as many as 40%) and this poses a question about why people who are committed to evidence-based propositions in their work would simultaneously accept faith-based beliefs in their private lives (MacPherson & Kelly, 2011). The MacPherson and Kelly study found that although scientists as a group were generally less religious than non-scientists, they also scored higher in creativity and magical thinking. Additionally, among these scientists, religious belief, creativity, and magical thinking were all positively correlated. This suggests that among scientists there is a subgroup that is religious, creative and prone to unconventional views of reality. The authors suggested that religious and paranormal beliefs may be linked to creativity and that these may form part of a broader psychological dimension known as transliminality. Transliminality involves the crossing of psychological material into and out of awareness. People who are high in transliminality may feel more able to accommodate the apparent tension between rational and non-rational ideas. This notion seems similar to that of Wolfradt et al. who proposed that the “complementary” thinking style fosters loose associations that support magical thinking associated with paranormal beliefs.

Future research could examine what effects experimental manipulations of analytical thinking might have on complementary thinkers, such as religious scientists. It seems possible that in order to decrease the religious beliefs of a complementary thinker it would be necessary not only to strengthen their analytical thinking but to weaken their reliance on intuition. Such research might be of particular interest for example to those who are concerned about the increasing prominence of non-evidence based treatments such as alternative medicine in educational institutions


Gervais, W. M., & Norenzayan, A. (2012). Analytic Thinking Promotes Religious Disbelief. Science, 336(6080), 493-496. doi: 10.1126/science.1215647

MacPherson, J. S., & Kelly, S. W. (2011). Creativity and positive schizotypy influence the conflict between science and religion. Personality and Individual Differences, 50(4), 446-450. doi: 10.1016/j.paid.2010.11.002

Pennycook, G., Cheyne, J. A., Seli, P., Koehler, D. J., & Fugelsang, J. A. (2012). Analytic cognitive style predicts religious and paranormal belief. Cognition, 123(3), 335-346. doi: 10.1016/j.cognition.2012.03.003

Shenhav, A., Rand, D., & Greene, J. (2011). Divine intuition: Cognitive style influences belief in God. Journal of Experimental Psychology: General. doi: 10.1037/a0025391

Wolfradt, U., Oubaid, V., Straube, E. R., Bischoff, N., & Mischo, J. (1999). Thinking styles, schizotypal traits and anomalous experiences. Personality and Individual Differences, 27(5), 821-830. doi: 10.1016/s0191-8869(99)00031-8

This post was previously published on my personal blog Eye on Psych

Be sure to read the following responses to this post by our bloggers:

Most Recent Posts from Unique—Like Everybody Else

The Mating Strategies of Extraverts

Extraversion is linked to both short and long-term mating

Does Watching TV Encourage Narcissism?

Research links TV with narcissism, but causality remains unclear

Psychedelic Drugs and the Nature of Personality Change

LSD use may increase openness to experience even in regular users