Cognitive Daily

Who believes in magic?

Babies love to play peek-a-boo. This simple game can entertain them for hours, even if all you do is hide your face behind your hands. Part of the reason is that for babies, it is really something of a surprise that you return. For most of their first year, babies don’t understand that objects exist beyond their view–out of sight is indeed out of mind. Sometime during the first year of life, children develop “object permanence”–they learn that objects still exist even when they are out of view.

In the 75 years since the Swiss psychologist Jean Piaget first observed this phenomenon, there has been a tremendous amount of additional research on object permanence. You might think the topic has become exhausted, but Eugene Subbotsky of Lancaster University has come up with an innovative series of experiments exploring how the concept of object permanence relates to our understanding of “magic” (“The Permanence of Mental Objects: Testing Magical Thinking on Perceived and Imaginary Realities,” Developmental Psychology, 2005).

One test of object permanence has been to use sleight of hand to make an object appear to change into another object—for example, putting a toy rabbit into a box, and then pulling out a toy dog. Three- to five-year-olds will identify this as “magic,” but what they really mean is that it’s a “trick”—they know that it’s impossible to change a rabbit into a dog, and so therefore the experimenter must have somehow deceived them.

Subbotsky wanted to know how the concept of object permanence related to the concept of magic. Do kids ever actually believe in magic, and at what age do those beliefs change? If older kids believe in magic, then does that mean they don’t have a complete understanding of object permanence? Is there ever a situation where adults continue to believe in magic? The experiments Subbotsky designed to answer these questions are so clever and elegant that Cognitive Daily will spend the next two days discussing them.

Subbotsky’s experiments involved a couple of innovations. First, he “transformed” objects using a particularly effective device: a small, wooden box that participants could hold in their hands while the transformation occurred. Participants were even allowed to inspect the box to make sure it appeared to be an ordinary device with no trap doors or hidden compartments. Of course, the box really did have a trap door inside to initiate the transformations, but it was hidden well enough that the participants never found it. Second, he administered the experiment in two different ways—by claiming that the transformation truly was magical, and by simply showing the device in action and letting participants arrive at their own conclusions about how it worked.

In experiment 1, Subbotsky tested 6-year-olds, 9-year-olds, and adults, showing them a blank piece of paper, then asking them to place it into the box. When the box was opened, a picture of a fish had “appeared” on the piece of paper. In every case, most 6-year-olds believed the transformation was magical. Even though these kids would have easily passed the Piagetian object permanence test, they seemed to believe permanence does not always apply. The 9-year-olds tended to believe it was a trick when not presented as “magic,” but when the experimenter acted as if he was casting a magical spell, like the 6-year-olds, the older children believed the transformation was magical. Adults were not so easily fooled, and believed the transformation was the result of trickery in both cases.

Subbotsky also added another wrinkle to the experiment: he asked participants to conduct the same experiment in their imaginations. Instead of putting a real piece of paper in a real box, he asked them to imagine doing it, and imagine the drawing of the fish appearing. The results were the same as in the “real” experiment.

Why didn’t the older kids see the non-magical version of the task as a magical transformation, and why was there no difference between the imaginary task and the real one? Subbotsky speculated that the imaginary objects were too real, and so in expermint 2 he replacing them with truly fictional objects. Instead of a picture appearing on paper, participants were asked to imagine a 1-inch long dog with wings transforming into an equally tiny cat with a fish’s tail.

He also repeated the trick with the magic box, but instead of a drawing appearing on the page, a card in the shape of a small green rabbit “transformed” into a large orange fish.

Here is a chart comparing the results of the first two experiments. First, the results of the real experiment, where the rabbit changed into a fish.


Scores here are charted by how “permanent” the participants thought the objects in the experiments were. If they believed that the experimenter was simply tricking them, then the scores were higher, and if they believed it was an example of real magic, the scores were lower. Interestingly, everyone was less likely to believe the second experiment was magical. Even the six-year-olds understood it to be a trick.

In the imaginary condition, where participants conducted the trick in their imaginations on fantastic creatures, the results were even more surprising:


Now the children were more likely to believe that the transformation was just a trick, and adults seemed to suspend disbelief and consider the transformation to be an example of true magic. Adults seem to have different rules for the fictional world of imagination and the real world, while children see the fictional world operating under similar rules to the real world.

We’ll continue our discussion of Subbotsky’s work tomorrow.


  1. [...] esterday we discussed the difference between children’s and adults’ beliefs in magic. Today we will continue that discussion, with two more experiments from the same article by Eugene Subbotsk [...]

  2. #2 bryan
    May 10, 2005

    I seem to remember reading about 3-4 years ago that a series of more in depth experiments had shown that babies did in fact have an expectation that things where still there, and that the enjoyment of the peekaboo game had to do more with enjoyment of interaction with a friendly adult.