Researchers have often puzzled over why people believe in the religious and paranormal. Neither of these beliefs has scientific evidence yet the vast majority of people still believe in them. It would seem to be a mystery why so many would believe in things so suspect. Tackling this problem head-on, neuropsychologist Marjaana Lindeman (University of Helsinki) and colleagues look for the cognitive biases that underlie supernatural beliefs, and they found that ontological confusions best predicted religious and paranormal belief.
Cognitive biases that warranted testing, according to the researchers, included empathy, anthropomorphism, and teleological bias. The emotion empathy, when so strong that it overrides the brain’s ability to systematize the world, could lead to belief in the supernatural. Next, “anthropomorphism” denotes the tendency to attribute human characteristics to nonhuman things. While anthropomorphism does not always involve the supernatural (for example, believing that your pet dog loves you) it could lead to belief in supernatural. Finally, teleological bias occurs when one reasons by looking at the purpose of something (for example, the purpose of the heart is a pump blood), and this “in order to” style of reasoning is called “promiscuous teleology.”
With these terms in mind, the researchers made four hypotheses. First, that women have more empathy and therefore more supernatural beliefs than men. Second, that mentalizing abilities will predict promiscuous teleology in ontological confusions. Third, that, in turn, promiscuous teleology and ontological confusions will predict religious and paranormal beliefs and belief in supernatural purposes. Fourth and finally, that hyper-empathizing will predict supernatural beliefs.
To test their hypotheses, the authors recruited 2,789 Finnish participants through open Internet discussion forums, student mailing lists, and from a pool of those who had expressed interests in the authors’ studies. Each volunteer would undergo a series of tests. The first series of tests measured empathy. The Pictorial Empathy Test (PET) presented pictures of men, women, and children feeling a range of emotions, mainly negative, and then asked the participants to rate how “touching” they found the photograph. Next the Adult Reading the Mind in the Eyes test likewise showed participants pictures, only this time asked them to choose the correct emotion word that best fits the picture. The participants then completed a shortened version of the Empathy Quotient scale (EQ), which tests for social skills, emotional reactivity, and empathy (one item includes “I can sense if I am intruding, even if the other person does not tell me’’). To calculate hyper-empathy, the researchers subtracted the SQ score from the EQ score. The SQ score, or Systemizing Quotient scale, measures how well people can mentally visualize the structures of things (a sample item includes, “I can easily visualize how the motorways in my region link up”). Finally, the Tromsø Social Intelligence Scale measured the social intelligence of all participants.
With empathy sufficiently measured, the researchers also had to measure knowledge confusion, teleological bias, and supernatural belief. The statements from the Core Knowledge Confusions scale concerning ontological confusions involved seeing whether participants would mind mixing up the properties of things inappropriately, for example, by giving animate properties to inanimate objects, and, if so, to what extent this mixing is metaphorical or literal. Then participants have to evaluate a series of explanations as good or bad and in the process their teleological bias would shine through (examples include “Earthworms tunnel underground to aerate the soil” and “Mosses form around rocks to stop soil erosion”). Finally, in order to measure belief in the supernatural, all the participants completed the Supernatural Beliefs Scale, which includes paranormal and religious beliefs.
The researchers found support for some but not all of their hypotheses. In support of the first hypothesis, women indeed had more religious beliefs, paranormal beliefs, and beliefs concerning supernatural purpose. They also empathized more and exhibited more hyper-empathy compared to men. However, the second hypothesis found little support (the path from empathy to ontological confusion was simply not significant). The third hypothesis had moderate support: ontological confusion and promiscuous teleology predicted all three types of supernatural belief (religious, paranormal, and supernatural purpose). Lastly, the results gave some support for the fourth hypothesis – while the researchers found no connection between hyper-mentalizing and ontological confusion, there was a significant relationship between hyper-mentalizing and promiscuous teleology.
The researchers conclude that “Considering the results overall, the present evidence provides strong support for the view that ontological biases predict supernatural beliefs but little support for the prediction that cognitive biases mediate the effect of mentalizing abilities on supernatural beliefs.” In other words, a strong relationship exists between ontological confusion and supernatural beliefs, but the reason for this may not have to do with cognitive biases. So some progress has been in solving the mystery of religious belief, hopefully without bias of its own.
For more see, “Ontological confusions but not mentalizing abilities predict religious belief, paranormal belief, and belief in supernatural purpose” in the journal Cognition.