Children Are Not “Natural” Scientists

A pernicious myth, repeated with good intentions in many places and by many people, is that children are natural scientists. They are born with something that gets beaten or worn out of them by bad teachers, bad schools, bad educational practices, and then must relearn what it means to be a scientist later in life. Like many myths, there’s a mixture of truth and falsehood, but ultimately the myth is damaging and leads us into bad habits of thought.

After all, what we really mean when we say “children are natural scientists” is that kids are curious about the world and open to learning about it. So, last year’s Scientific American article “More Than Child’s Play: Ability to Think Scientifically Declines as Kids Grow Up” by Sharon Begley, both intrigued and bothered me. “Thinking like a researcher” is not the same thing as a natural curiosity and mental plasticity — scientific research is very much a learned skill, in my experience, but I admit to being entirely ignorant of child development, so maybe I was missing out on something.

However, the great thing about the internet is if you wait long enough (in that case, less than two days!), an expert will step up. Marie-Claire Shanahan in her essay “Students Don’t Lose Their Ability to Think Scientifically” points out that there are indeed two things going on: the specific idea of experimentation, which young children are pretty good at, and the more general, abstract idea of controlled research. As she writes,

It is one thing to measure the length of a particular piece of string, quite another to conceive of length as a general property that can be measured or manipulated in any object. This especially true because it is also somewhat arbitrary, requiring the person doing the experiment to choose an operational definition (e.g., by defining length as the measurement of the longest side). There is no concrete thing called length. It is an abstract word that describes a type of measurement.

Marie-Claire’s follow-up post goes even farther:

[S]cience isn’t just a grown up version of a child’s curiosity. While kids have the fertile beginnings, becoming a scientist requires that they learn and skillfully practice many abstract skills that are far from intuitive. When students struggle with scientific thinking later in life it isn’t because they have unlearned or lost the ability, it’s because they (for any number of reasons) didn’t get to take the next steps to developing those skills and understandings.

I’m going to tentatively say that none of us are “born scientists”. We have the potential to become scientific thinkers and researchers, and some of us will have an easier go of it than others (for reasons springing both from nature and nurture). The job of an educator is to feed the curiosity, use the plasticity, and help the student build scientific intuition, to borrow a phrase my college advisor often used. It may seem paradoxical to talk about building intuition, but it seems to be a useful concept: we can learn to think in certain ways, especially if we start early. We humans aren’t naturally scientific or skeptical thinkers: we often get the wrong end of the stick on basic notions like correlation/causation, randomness, probability, and the like.

Think of it this way: say you toss a coin fives times in a row, and it comes up heads all five times. Our natural inclination is to think that the coin is more likely to come up tails on the sixth toss, but it’s not—the chance is still 50% (well, approximately, since coins aren’t perfectly symmetrical). Successive throws of the coin don’t depend on the previous toss: if you repeat the trial 50 million times, the 50,000,001st trial will still have a 50/50 chance of landing on tails, no matter how many heads or tails preceded it. Lest you think this only relates to coins, think the couples choosing to have another child with the specific hope of giving birth to a boy or girl, having birthed several children of the opposite gender previously. However, the sex of a specific baby doesn’t depend on the sex of its older siblings. (It’s true that some factors can skew the gender of a child, but based on my limited understanding, those are undesirable to subject yourself to.) The intuition has the feeling of correctness—after all, if you look at the population at large or a huge number of coin tosses, you’ll see a rough 50/50 division of girls and boys or heads and tails. However, each individual toss or birth is itself effectively random.

This kind of woolly thinking is as true for professional scientists as for non-scientists, but training can help. Assuming that scientific training automatically makes you immune is a common pitfall, alas (just as being brilliant doesn’t keep you from being a sexist deep-sea grade douchecanoe). The key in all of this is to remain skeptical even of your own thinking, not to trust it without evidence—and that’s something that can be learned. Another recent example is the much-touted success by statistician Nate Silver in predicting the outcome of the 2012 United States Presidential election. Silver’s basic premise was simple: polls should provide enough information to know what the voters will decide. That’s it! The rest was statistical techniques that can be learned; I know what they are, in fact, and have used them in modeling white dwarf binary systems. But many political reporters acted as though he had done something new and phenomenal, because their methodology is based on an intuitive reading of the national mood, or some similar set of ideas—in contrast to their own stated beliefs in polling.

One of my favorite memories from my all-too-brief tenure as a planetarium director was the group of five or six young kids who came up to me after a show to ask, “Are you a scientist?” in an awed voice. Yes, I am. And they can be too. It’s something they can learn, and grow into, and flourish.

[Adapted and enlarged from an earlier post. You know the drill: writing my book, not much time for blogging.]

13 responses to “Children Are Not “Natural” Scientists”

  1. “Think of it this way: say you toss a coin fives times in a row, and it comes up heads all five times. Our natural inclination is to think that the coin is more likely to come up tails on the sixth toss, but it’s not—the chance is still 50% (well, approximately, since coins aren’t perfectly symmetrical).”

    This is true only if you have prior and independent knowledge that it’s a fair coin. Otherwise, it’s a natural Bayesian thing to do to update your prior belief with the growing suspicion that the coin may be biased, in which case the probability of heads is > 50 %.

    1. True enough.

      Or, one might say, not quite true enough. The problem is that this statement cannot be understood before one has acquired a considerable amount of knowledge that is thoroughly non-, even counter-, intuitive.

      For a long time these nitpicking distinctions didn’t matter, in any obvious practically important way. But now the possibility of doing science depends on this stuff. Hence, for me, it’s a case in point for the posting.

  2. There are in addition many reasons why humans are simply not born good scientists. We have a lot of cognitive biases and fall prey to social influence. That’s why scientists use a lot of techniques to avoid they fool themselves, think double-blind trials, peer review, statistical analysis and so on. You’re not born with that and it’s not intuitive.

  3. Fascinating and well-written post! And I agree whole-heartedly. You know that I have 2 kids, and you probably remember my boy (who we now know to have Asperger’s, so maybe he’s not a good example). His favorite (ad nauseum) thing to say is, “Nothing can be true unless it is proven to be 100% true”. I really do believe he may find his destiny in science one day, and he is certainly a born explorer and questioner of the establishment. But I do wonder how much of this is nurture, as well. Since we are not religious people and are raising our kids in a critical-thinking household, I wonder if my kids have a better chance of becoming critical thinkers than most? I am not saying here that “my kids are better than yours”. I am only saying it is a difference in how we teach our kids to view the world. I agree that children are born to be curious. That is the job of children – to sort out the big confusing world. But how much of that natural curiosity is squashed by parenting styles that do not leave room for that curiosity to grow? Or that does not leave room for questioning established thinking? I realize I have gone off onto a tangent here that was not the focus of your post. Suffice it to say that I think some children have a better chance to become scientists than others.

    1. Newsjudy? I’d say your child is going to develop into another Einstein. I’m being facetious, of course. “It’s not true until it’s proven one hundred percent!!!!” is an enormous pitfall to learning. I pray, and HOPE! your children can get beyond that.

  4. […] still a far cry from being natural-born “scientists”. To suggest otherwise, warns real-life, classically trained astrophysicist Matthew Francis, is to cultivate a pernicious myth which, “like many myths, [is] a mixture of truth and […]

  5. […] « Children Are Not “Natural” Scientists […]

  6. While I think your arguments are sound for the point you’re making, you’re also not engaging with the actual research (e.g. and points made by those who are making the claim, who are usually arguing that methods of teaching children can *encourage* curiosity and experimentation, or they can *discourage* it, and that we should use the former. In other words, you’re attacking a straw man.

    1. OK, I see that a few links in to what you’re responding to does come from the Cook paper. I revise my criticism from “straw man” to “taking deGrasse Tyson’s statement too literally.”

  7. […] Now, we have the dissenting view, from Matthew Francis at Galileo’s Pendulum, in his post “Children Are Not “Natural” Scientists“: […]

  8. The coin example is rather odd; we also naturally believe in runs of good luck (Don’t jinx it, it’ll land heads again!) and bad luck (This is the fifth time, I guess I’m just having a losing streak.) and perhaps most naturally of all would suspect the coin to be biased for not doing what we expected.

    I think the biggest sign that children (and adults) are not natural scientists is their willingness to assert that they are right and to make up explanations for when ever they are proved wrong. There is nothing more obnoxious and anti-science than someone who argues against all evidence to the contrary that something isn’t their fault.

  9. I agree but I don’t think this goes far enough. I would say that most human children are effectively doomed to a life of ignorance.

%d bloggers like this: