I suggest that we not talk about the brains of our students but rather their behavior. After all, if we figured out some way to improve their brains—increase the volume of specific regions, say, or the number of interconnections—but we saw no change in their ability to succeed at their actual schoolwork, we would not be satisfied.

This distinction may sound like a matter of semantics, but there is an important practical implication. Focusing on the brain implies that once a brain is in good shape, it will be better equipped to take on any task that comes along. There is probably some truth in that statement—some thinking skills are quite general. But in practice, enhancing them has proved difficult.

In the past decade or so researchers have experimented with various exercises meant to increase one such skill, namely the capacity of working memory. Working memory is the mental space you use to keep several things in mind at once—say, numbers such as 38 and 16—and to manipulate them, perhaps by multiplying them together. As you might guess, people who can keep more things in working memory and who are more efficient at manipulating them tend to be better at reasoning.

But the effectiveness of working memory training is controversial. Everyone agrees that people get better at the specific tasks they practice, but it is less clear whether that skill transfers to other, unpracticed tasks. Learning how to calculate least common multiples, for example, does not make you better at math in general.

This specificity is especially pronounced early in training. When someone spends years working at certain types of problems, they do develop thinking skills that can be more flexibly deployed. For example, a professional historian who specializes in the Italian Renaissance can do a creditable job analyzing documents from the American Civil War. But even so, experience matters. Our historian's training has made the person good at thinking like a historian, not good at thinking generally. Or to put it another way, Stephen Hawking may be a very smart guy, but I would not suggest he coach the Chicago White Sox.

Rather than thinking about developing our students' brains, I suggest focusing on specific thinking skills. What is a good writer able to do? What are our expectations for mathematical thinking? We must define the abilities that go into our definition of competence in each domain and give students ample practice in honing them.

Question submitted by Lola Irele, London