It’s only a matter of time—in fact, they’ve already started cropping up—before reality-challenged individuals begin pontificating about what God could have possibly been so hot-and-bothered about to trigger last week’s devastating earthquake and tsunami in Japan. (Surely, if we were to ask Westboro Baptist Church members, it must have something to do with the gays.) But from a psychological perspective, what type of mind does it take to see unexpected natural events such as the horrifying scenes still unfolding in Japan as “signs” or “omens” related to human behaviors?

In the summer of 2005, my University of Arkansas colleague Becky Parker and I began the first experimental study to investigate the psychology underlying this strange phenomenon. In this experiment, published the following year in Developmental Psychology, we invited a group of three- to nine-year-old children into our lab and told them they were about to play a fun guessing game. It was a simple game in which each child was tested individually. The child was asked to go to the corner of the room and to cover his or her eyes before coming back and guessing which of two large boxes contained a hidden ball. All the child had to do was place a hand on the box that he or she believed contained the ball. A short time was allowed for the decision to be made but, importantly, during that time the children were allowed to change their mind at any time by moving their hand to the other box. The final answer on each of the four trials was reflected simply by where the child’s hand was when the experimenter said, “Time’s up!” Children who guessed right won a sticker prize.

In reality, the game was a little more complicated than this. There were secretly two balls, one in each box, and we had decided in advance whether the children were going to get it “right” or “wrong” on each of the four guessing trials. At the conclusion of each trial, the child was shown the contents of only one of the boxes. The other box remained closed. For example, for “wrong” guesses, only the unselected box was opened, and the child was told to look inside (“Aw, too bad. The ball was in the other box this time. See?”). Children who had been randomly assigned to the control condition were told that they had been successful on a random two of the four trials. Children assigned to the experimental condition received some additional information before starting the game. These children were told that there was a friendly magic princess in the room, “Princess Alice,” who had made herself invisible. We showed them a picture of Princess Alice hanging against the door inside the room (one that looked remarkably like Barbie), and we gave them the following information: “Princess Alice really likes you, and she’s going to help you play this game. She’s going to tell you, somehow, when you pick the wrong box.” We repeated this information right before each of the four trials, in case the children had forgotten.

For every child in the study, whether assigned to the standard control condition (“No Princess Alice”) or to the experimental condition (“Princess Alice”), we engineered the room such that a spontaneous and unexpected event would occur just as the child placed a hand on one of the boxes. For example, in one case, the picture of Princess Alice came crashing to the floor as soon as the child made a decision, and in another case a table lamp flickered on and off. (We didn’t have to consult with Industrial Light & Magic to rig these surprise events; we just arranged for an undergraduate student to lift a magnet on the other side of the door to make the picture fall, and we hid a remote control for the table lamp surreptitiously in the experimenter’s pocket.) The predictions were clear: if the children in the experimental condition interpreted the picture falling and the light flashing as a sign from Princess Alice that they had chosen the wrong box, they would move their hand to the other box.

What we found was rather surprising, even to us. Only the oldest children, the seven- to nine-year-olds, from the experimental (Princess Alice) condition, moved their hands to the other box in response to the unexpected events. By contrast, their same-aged peers from the control condition failed to move their hands. This finding told us that the explicit concept of a specific supernatural agent—likely acquired from and reinforced by cultural sources—is needed for people to see communicative messages in natural events. In other words, children, at least, don’t automatically infer meaning in natural events without first being primed somehow with the idea of an identifiable supernatural agent such as Princess Alice (or God, one’s dead mother, angels, etc.).

More curious, though, was the fact that the slightly younger children in the study, even those who had been told about Princess Alice, apparently failed to see any communicative message in the light-flashing or picture-falling events. These children kept their hands just where they were. When we asked them later why these things happened, these five- and six-year-olds said that Princess Alice had caused them, but they saw her as simply an eccentric, invisible woman running around the room knocking pictures off the wall and causing the lights to flicker. To them, Princess Alice was like a mischievous poltergeist with attention deficit disorder: she did things because she wanted to, and that’s that. One of these children answered that Princess Alice had knocked the picture off the wall because she thought it looked better on the ground. In other words, they completely failed to see her “behavior” as having any meaningful connection with the decision they had just made on the guessing game; they saw no “signs” there.

The youngest children in the study, the three- and four-year-olds in both conditions, only shrugged their shoulders or gave physical explanations for the events, such as the picture not being sticky enough to stay on the wall or the light being broken. Ironically, these youngest children were actually the most scientific of the bunch, perhaps because they interpreted “invisible” to mean simply “not present in the room” rather than “transparent.” Contrary to the common assumption that superstitious beliefs represent a childish mode of sloppy and undeveloped thinking, therefore, the ability to be superstitious actually demands some mental sophistication. At the very least, it’s an acquired cognitive skill.

Still, the real puzzle to our findings was to be found in the reactions of the five- and six-year-olds from the Princess Alice condition. Clearly they possessed the same understanding of invisibility as did the older children, because they also believed Princess Alice caused these spooky things to happen in the lab. Yet although we reminded these children repeatedly that Princess Alice would tell them, somehow, if they chose the wrong box, they failed to put two and two together. So what is the critical change between the ages of about six and seven that allows older children to perceive natural events as being communicative messages about their own behaviors (in this case, their choice of box) rather than simply the capricious, arbitrary actions of some invisible or otherwise supernatural entity?

The answer probably lies in the maturation of children’s theory-of-mind abilities in this critical period of brain development. Research by University of Salzburg psychologist Josef Perner, for instance, has revealed that it’s not until about the age of seven that children are first able to reason about “multiple orders” of mental states. This is the type of everyday, grown-up social cognition whereby theory of mind becomes effortlessly layered in complex, soap opera–style interactions with other people. Not only do we reason about what’s going on inside someone else’s head, but we also reason about what other people are reasoning is happening inside still other people’s heads! For example, in the everyday (nonsupernatural) social domain, one would need this kind of mature theory of mind to reason in the following manner:

“Jakob thinks that Adrienne doesn’t know I stole the jewels.”
Whereas a basic (“first-order”) theory of mind allows even a young preschooler to understand the first propositional clause in this statement, “Jakob thinks that . . . ,” it takes a somewhat more mature (“second-order”) theory of mind to fully comprehend the entire social scenario: “Jakob thinks that [Adrienne doesn’t know] . . .”

Most people can’t go much beyond four orders of mental-state reasoning (consider the Machiavellian complexities of, say, Leo Tolstoy’s novels), but studies show that the absolute maximum in adults hovers around seven orders of mental state. The important thing to note is that, owing to their still-developing theory-of-mind skills, children younger than seven years of age have great difficulty reasoning about multiple orders of mental states. Knowing this then helps us understand the surprising results from the Princess Alice experiment. To pass the test (move their hand) in response to the picture falling or the light flashing, the children essentially had to be reasoning in the following manner:

“Princess Alice knows that [I don’t know] where the ball is hidden.”

To interpret the events as communicative messages, as being about their choice on the guessing game, demands a sort of third-person perspective of the self’s actions: “What must this other entity, who is watching my behavior, think is happening inside my head?” The Princess Alice findings are important because they tell us that, before the age of seven, children’s minds aren’t quite cognitively ripe enough to allow them to be superstitious thinkers. The inner lives of slightly older children, by contrast, are drenched in symbolic meaning. One second-grader was even convinced that the bell in the nearby university clock tower was Princess Alice “talking” to him.

Princess Alice may not have the je ne sais quoi of Mother Mary or the fiery charisma of the Abrahamic God we’re all familiar with, but she’s arguably a sort of empirically constructed god-by-proxy in her own right. The point is, the same basic cognitive processes—namely, a mature theory of mind—are also involved in the believer’s sense of receiving divine guidance from these other members of the more popular holy family. When people ask God to give them a sign, they’re often at a standstill, a fork in the road, paralyzed in a critical moment of existential ambivalence. In such cases, our ears are pricked, our eyes widened, our thoughts ruminating on a particular problem—often “only God knows” what’s on our minds and the extent to which we’re struggling to make a decision. It’s not questions like whether we should choose a different box, but rather decisions such as these: Should I stay with this person or leave him? Should I risk everything, start all over in a new city, or stay here where I’m stifled and bored? Should I have another baby? Should I continue receiving harsh treatment for my disease, or should I just pack it in and call it a life? Just like the location of the hidden ball inside one of those two boxes, we’re convinced that there’s a right and a wrong answer to such important life questions. And for most of us, it’s God, not Princess Alice, who holds the privileged answers.


God doesn’t tell us the answers directly, of course. There’s no nod to the left, no telling elbow poke in our side or “psst” in our ear. Rather, many envision God, and other entities like Him, as encrypting strategic information in an almost infinite array of natural events: the prognostic stopping of a clock at a certain hour and time; the sudden shrieking of a hawk; an embarrassing blemish on our nose appearing on the eve of an important interview; a choice parking spot opening up at a crowded mall just as we pull around; an interesting stranger sitting next to us on a plane. The possibilities are endless. When the emotional climate is just right, there’s hardly a shape or form that “evidence” cannot assume. Our minds make meaning by disambiguating the meaningless.

This sign-reading tendency has a distinct and clear relationship with morality. When it comes to unexpected heartache and tragedy, our appetite for unraveling the meaning of these ambiguous “messages” can become ravenous. Misfortunes appear cryptic, symbolic; they seem clearly to be about our behaviors. Our minds restlessly gather up bits of the past as if they were important clues to what just happened. And no stone goes unturned. Nothing is too mundane or trivial; anything to settle our peripatetic thoughts from arriving at the unthinkable truth that there is no answer because there is no riddle, that life is life and that is that.


Text by Jesse Bering | Sunday, March 13, 2011 |

(Author’s note: Some of the foregoing material is excerpted, with edits, from my new book, The Belief Instinct: The Psychology of Souls, Destiny and the Meaning of Life.)