The Science of Superstition

No one is immune to magical thinking.

Robin Davey

A visitor once asked the Nobel Prize–winning physicist Niels Bohr whether he really believed that the horseshoe he’d hung at his country home was lucky. “Of course not,” Bohr said. “But I understand it’s lucky whether you believe in it or not.”

If Bohr couldn’t resist magical thinking, can anyone? One recent study found that even physicists, chemists, and geologists at MIT and other elite schools were instinctively inclined to attach a purpose to natural events. When the researchers subjected the scientists to time pressure (reasoning that this could expose a person’s uncensored biases), they were twice as likely to approve of statements such as “Trees produce oxygen so that animals can breathe” than they were when they had time to respond more deliberately [1]. Such bias may well be deep-seated: another recent study found that, regardless of their parents’ religiosity, 5-to-7-year-old children preferred explanations of events that involved lessons—like “Maggie’s house burned down to teach her not to play with fire anymore” [2].

Even atheists seem to fear a higher power. A study published last year found that self-identified nonbelievers began to sweat when reading aloud sentences asking God to do terrible things (“I dare God to make my parents drown”). Not only that, they stressed out just as much as believers did [3]. Belief in the soul also appears widespread. One psychologist found that among people who said that consciousness ends at death, a third nonetheless attributed ongoing thoughts and feelings to characters in a fictional story after those characters had died [4].

Expressions of subconscious religious belief seem to increase when we are reminded of our own mortality—as they say, there are no atheists in foxholes. In one study, writing about death increased subjects’ implicit associations between words for supernatural entities (God, soul, hell) and synonyms for real (true, factual). The effect was equally strong in Christians and in people who described themselves as nonreligious [5].

Magical thinking is not just a result of ignorance or indoctrination—it appears to be a side effect of normal, socially adaptive thinking: we attribute intentions to the natural world in much the same way that we attribute intentions to other people. Indeed, a recent paper from a lab at the University of British Columbia reported that the better study participants were at reading others, the more strongly they believed in God, the paranormal, and the notion that life has a purpose [6]. Meanwhile, one of the few true avenues to atheism may be autism. The same lab found that the more autistic traits a person had, the less likely he or she was to believe in God [7].

Fear is another driver of irrationality. In a British study, students imagined an encounter with a self-professed witch who offered to cast an evil spell on them. About half said a scientist should accept the hex without concern. Yet each of them said that, personally, they’d decline the offer [8].


The Studies:

[1] Kelemen et al., “Professional Physical Scientists Display Tenacious Teleological Tendencies” (Journal of Experimental Psychology: General, Nov. 2013)

[2] Banerjee and Bloom, “ ‘Everything Happens for a Reason’ ” (Child Development, published online Oct. 2014)

[3] Lindeman et al., “Atheists Become Emotionally Aroused When Daring God to Do Terrible Things” (International Journal for the Psychology of Religion, 2014)

[4] Bering, “Intuitive Conceptions of Dead Agents’ Minds” (Journal of Cognition and Culture, 2002)

[5] Jong et al., “Foxhole Atheism, Revisited” (Journal of Experimental Social Psychology, Sept. 2012)

[6] Willard and Norenzayan, “Cognitive Biases Explain Religious Belief, Paranormal Belief, and Belief in Life’s Purpose” (Cognition, Nov. 2013)

[7] Norenzayan et al., “Mentalizing Deficits Constrain Belief in a Personal God” (plos One, 2012)

[8] Subbotsky, “The Permanence of Mental Objects” (Developmental Psychology, March 2005)