On this page
A child asks why the sky is blue. The quickest response is the answer. A slower one, "What do you think?", often teaches more. This guide covers what research says about curiosity and learning, and gives you a set of questions to use at home, with examples for younger children and teenagers.
Does curiosity actually help children learn?
Several lines of research suggest it does.
In a study by Kang and colleagues (Psychological Science, 2009), adults read trivia questions while their brains were scanned. The more curious they felt about a question, the more activity appeared in brain regions linked to anticipated reward. In follow-up experiments, people spent more of their limited tokens or waiting time to learn answers they were curious about, and higher curiosity went with better recall of surprising answers one to two weeks later.
Gruber, Gelman and Ranganath (Neuron, 2014) found that people remembered information they were curious about better, both immediately and a day later. They also remembered unrelated material shown to them while they were in a curious state.
For younger children, Shah and colleagues (Pediatric Research, 2018) analysed 6,200 children in a large US study and found that greater curiosity, as reported by parents, was associated with higher kindergarten reading and maths scores, after accounting for self-control and other factors. The link was stronger for children from lower socioeconomic backgrounds. This is a correlation, not proof that curiosity causes the gains.
Why asking beats telling
When you answer immediately, the child's question closes. When you ask back, the child has to generate a guess, and a guess creates a gap they now want to fill.
Two classroom findings help here. Mary Budd Rowe's work on wait time (Journal of Teacher Education, 1986) observed that teachers typically wait less than one second after asking a question before speaking again; her review discusses pausing longer, both after a question and after a student answers. And a large review of study techniques by Dunlosky and colleagues (2013) rated elaborative interrogation, prompting learners to explain why a fact is true, as of moderate utility: promising, though not yet tested enough in real classrooms.
The question ladder: what to ask instead of answering
Use these in order. Stop when the child is happily thinking.
"What do you think?" Invite a guess. Any guess.
"Why do you think that?" Ask for a reason, not a right answer.
"How could we find out?" Turn the question into a small test or a search together.
"What would happen if...?" Change one thing and predict.
"Were you surprised?" After finding out, compare the result with the guess.
If you do not know the answer either, say so. "I don't know, let's find out" models curiosity better than a confident guess.
Examples by age
Ages 6 to 10
Child: "Why does ice float?" You: "What do you think? Does everything float?" Then test a few things in a bowl of water.
Child: "Where does the rain go?" You: "Where have you seen water after rain? Let's watch the drain next time."
Ages 11 to 14
Child: "Why do we see lightning before thunder?" You: "What travels faster, do you think: light or sound? How could we estimate how far away the storm is?"
Child: "Why does a pressure cooker cook faster?" You: "What happens to the boiling point when the lid is locked? Why might that matter?"
Ages 15 to 18
Teen: "Why do I need integration?" You: "Where in your physics book does an area under a graph show up? What does it mean there?"
Teen: "This chemistry reaction makes no sense." You: "Which step surprised you? What would you have predicted?"
For teenagers, the habit of asking "why" is close to self-explanation, which our article on the self-explanation effect covers in depth.
Make wrong guesses safe
Curiosity dies fastest when guesses are mocked. If a guess is wrong, ask what made it seem right. Our piece on why mistakes help you learn explains why a confident wrong guess, once corrected, is often remembered better.
Some questions are easier to explore by seeing them. "What happens to a pendulum if the string is longer?" can be tried in the Fermi Simulation Lab, which has interactive simulations for ideas that are easier to see than to read. Ask for a prediction first, then run it.
