Mixing Memory

It’s now clear that by age 3, children have a pretty sophisticated theory of mind, which includes an understanding of the limits of the causal powers of thought. They know that thoughts cause behaviors and other thoughts, but they’re also aware that simply thinking about something can’t affect it. Except, according to a recent paper by Woolley, Browne, and Boerger(1), when it comes to wishing. They cite results from a study by Vikan and Clausen(2) in which 96% of 4 to 6-year old children believed that their wishes could affect others. Which raises an interesting question: when do children believe wishes work?

In an earlier paper, Woolley et al.(3) found that young children treated wishing as magical. This suggests that wishing may not be subject to the constraints of ordinary causation. However, following Boyer’s(4) finding that people tend to treat magic as subject to ordinary causal constraints, Woolley, Browne, and Boerger (WB&B from now on) argue that some knowledge of constraints on ordinary causation may influence children’s reasoning about wishing. They list three constraints that cross-cultural research has found in people’s reasoning about ordinary causation, both mental and physical (wishing, it would seem, is a bit of both):

  • Priority: Causes occur before effects.
  • Consistence: The effect should be consistent with the attributes of the cause.
  • Exclusivity: People are less likely to infer that something is cause of a particular effect if there are other potential causal explanations for that effect.

To test whether these constraints had an affect on children’s thinking about wishing, WB&B conducted a fairly straightforward experiment. After asking children between the ages of 3 and 7 (half 3-4 and half 5-7), one of the experimenters, the wisher, was introduced as someone with a lot of things to wish for. In front of the wisher was a line of 10 different boxes.There were 5 different conditions. The first two types were “honored” and “failed” trials. These always involved showing the child that the box was empty, after which the wisher would wish for a particular object to appear in the box. The box would then be opened again, and the object would be inside (“honored” condition) or not (“failed” condition). Notice that these two conditions involve the presence of all three constraints on causality: priority, because the wish occurred before the object appeared; consistency, because the object that appears is the one that was wished; and exclusivity, because the child did not observe any other potential cause of the presence of the object in the box.

The next three types of trials each involved a violation of one of the three causal constraints. In the “violation of priority” trials, the wisher left the room before making a wish, and while he or she was gone, one of the other experimenters opened the box and showed the child that an item was in it. When the wisher got back, he or she wished for that item, and the child was again shown that it was in the box. In the “violation of consistency” trials, the the item that appeared in the box after the wish was different from the one wished for. In the “violation of exclusivity” trials, as the wisher wished for an item, the experimenter placed that item in the box as the child watched. After each trial (each child observed 2 of each type), the children were asked how they thought the object got in the box, except on the “failed” trials, in which they were asked, “What happened?”

To analyze the data, WB&B computed a “wish” score for each trial (other than the “failed” trials), with 1 = the wish caused the object to appear and 0 = the wish didn’t cause it to appear. As you might expect, WB&B found that the older children (5-7) were less likely to think that the wishes had come true (56%) than the younger children (3-4, 72%), across all 5 types of trials. Interestingly, they also found that girls were more likely to believe the wishes had come true (71%) than boys (58%). They don’t offer an explanation for the gender difference, noting only that it has not been observed before in studies of wishing. So it may just be a fluke, but it makes for an interesting topic for future research. The important question though, is do children’s answers in the violation trials differ from the “honored” trials? The results for each type of trial except the failed trials are shown in this graph:


As you can see, children’s “wish scores,” which amount to a percentage of trials on which they believed the wish caused the object to appear, were highest in the “honored” trials, which involved the honoring of all three causal constraints. Violations of exclusivity had the biggest effect on wish attribution, with children believing that the wish caused the object to appear on only 29% of those trials. Violations of priority also produced results significantly different from the honored trials, with children attributing the object’s presence to the wish 68% of the time. Violations of consistency had little effect on children’s wish attribution (75%). So it seems that when it comes to wishes coming true, children are most sensitive to violations of exclusivity, and somewhat sensitive to violations of priority, but don’t really care about violations of consistency (in this case, the wrong objects showing up in the box). Importantly, in two follow-up experiments, WB&B found that this pattern of results mirrored the pattern for children’s reading about physical (i.e., non-magical) causation. That is, for physical causes, children were most sensitive to violations of exclusivity, less sensitive to violations of priority, and pretty much insensitive to violations of consistency. So it appears that children reasoning about wishes is pretty much like their reasoning about any other sort of causation.

In summary, children’s reasoning about the causal efficacy of wishes is pretty much the same as their reasoning about causes in general. They understand that if there are alternative explanations, wishes probably aren’t the cause, and if the wishes come after the wished-for effect, then the wish is less likely to be the cause of that effect. However, they have no problem believing that wishes or physical causes produce inconsistent effects. That is, if I wish for a puppy, and get a kitten, children are perfectly willing to believe that my wish caused me to get a kitten, even though that’s not what I’d wished for. Apparently wishes are inaccurate.

1Woolley, J.D., Browne, C.A., & Boerger, E.A. (2006). Constraints on children’s judgments of magical causality. Journal of Cognition and Development, 7(2), 253-277.
2Vikan, A., & Clausen, S. E. (1993). Freud, Piaget, or neither? Beliefs in controlling others by wishful thinking and magical behavior in young children. The Journal of Genetic Psychology, 154, 297-314.
3Woolley, J. D., Phelps, K. E., Davis, D. L., & Mandell, D. J. (1999). Where theories of mind meet magic: The development of children’s beliefs about wishing. Child Development, 70, 571-587.
4Boyer, P. (1994). The Naturalness of Religious Ideas: A Cognitive Theory of Religion. Berkeley: University of California Press.


  1. #1 Flaky
    June 26, 2007

    What about the “failed” cases?

  2. #2 David Harmon
    June 26, 2007

    Sounds pretty sensible to me!

  3. #3 Chris
    June 26, 2007

    Flaky, they didn’t include failed trials in any of the analyses, because they were really just meant to make sure the children weren’t just saying, “Yes” or “No” all the time to objects showing up in the box.

  4. #4 Kevembuangga
    June 28, 2007

    … but they’re also aware that simply thinking about something can’t affect it. Except, according to a recent paper by Woolley, Browne, and Boerger(1), when it comes to wishing.

    That would be the interesting point to investigate.
    How is wishing felt as different from thinking!

  5. #5 Cheryl Browne
    July 30, 2007

    Hey, this is Cheryl (Browne), one of the authors. Just wanted to say that I like Kevembuangga’s idea to investigate how children’s ideas about wishing differ from “plain” thinking. There’s some research on this by Woolley with maybe D. Davis, M. Bruell, or K. Phelps, but it could go further. I know they asked questions like “Who can wish?”, what one has to do to make a wish (e.g., close your eyes, say certain words, etc.). I wonder if “wishing” is thought of as something close to praying (or like casting spells in some cultures) — e.g., it has to include certain conventions, like closing your eyes and putting your hands together, sign-ons and -offs, like “Dear God” and “Amen,” etc. , so that it would be a certain sub-type of thinking in which speech is used for a ritualistic act (or a “speech-act”) that has some communicative or generative purpose. Hmmm…