A research team at MIT, led by Elizabeth Bonawitz and Laura Schulz, gave a group of four-year-olds a custom toy with four hidden functions. Pull a yellow tube and it squeaks. Press a button inside a blue tube and it lights up. Press a yellow pad and it plays music notes. Look through two black tubes and you see your own face reflected back at you in a mirror.
Half the kids saw the experimenter pull the yellow tube and make the squeak, then she handed the toy over. The other half got the toy without a demonstration. Both groups played for about the same amount of time, around two minutes. The difference was what they did during play. The kids who saw the demonstration spent two-thirds of their time on the squeaker and discovered, on average, less than one of the three other functions. The kids who got no demonstration spent half their time on the squeaker and discovered, on average, one to two of the others.
This is the central finding of a literature parents almost never get told about. When you teach a child how something works, you also teach them that this is what it does. Their exploration narrows. They fixate on the function you showed them. The demonstration does its job efficiently. The cost is the part of the morning where the kid would have found more on their own.
A 2022 systematic review and meta-analysis at the University of Cambridge pulled together thirty-nine intervention studies on what happens when adults set up an activity, hold a learning goal in mind, and then let the child drive. Guided play, the researchers call it. The structure is loose. The adult is engaged. The child decides what to try, what to repeat, what to abandon. The Cambridge team found that guided play outperformed direct instruction on early math, shape knowledge, and task switching, and outperformed free play on spatial vocabulary. For other outcomes -- receptive vocabulary, behavior regulation, inhibitory control -- the pooled effects were unclear. The dose mattered: the strongest effects appeared in studies with mid-range exposure, roughly six to twenty sessions, not endless.
Translation: set up the conditions, step back, and ask a question instead of giving an answer.
The grocery store is a classroom. Not because you should turn it into one. Because it already is one and you have been doing the teaching whether you meant to or not. The aisle is geometry. The receipt is arithmetic. The label is reading. The “do you think this is enough or do we need two” is probability and estimation and self-trust all at once. You do not have to make the trip educational. Just stop closing off the part that already is.
The park is a classroom. A stick is a lever. A puddle is hydraulics. Climbing the wrong tree is risk assessment. The kid who finds out that this branch holds and that one does not has done physical science in a way no worksheet replicates.
The kitchen is a classroom. Measuring cups are fractions. The egg is biology. The flour-water ratio is what every recipe in human history has been arguing about. Cooking together is the original guided play, and most of us have stopped doing it because it takes longer and makes more mess than ordering DoorDash.
The instruction reflex is the part of you that wants to jump in. Demonstrate. Correct. Narrate. Explain. Tell them the right way before they find a wrong way. That reflex is even louder for parents of complicated kids where every interaction can feel like a missed window. The pull to fill the silence with teaching is real. The cost of filling it is what the toy experiment shows.
Set up. Step back. Ask.
This is not the same as walking away. Guided play is guided. The adult is in the room, paying attention, putting pieces in reach, asking the question at the right moment, taking the question seriously when it comes back. The difference between guided play and free play is the adult mind in the room. The difference between guided play and direct instruction is who is driving.
Direct instruction has its uses, and there are things that must be taught directly. Phonics. Multiplication tables. Swimming. The cross-walk light. Some skills are too high-stakes or too convention-bound to discover through trial and error. Use direct instruction for those. Use guided play for everything else, which turns out to be most of childhood.
Alison Gopnik has a framing that has been useful in our house and is worth borrowing. The carpenter sculpts a piece of wood into a known thing. The gardener creates the conditions and lets the plant become what it is. Most of modern parenting culture is set up like a workshop. Most of childhood development needs a garden.
The practical move is small enough to try this week. On the next ordinary errand, set up one moment. The cereal aisle. The walk to the car. The dishwasher loading. Ask one question instead of giving one instruction. “Which one do you think is bigger?” “Where do you think this goes?” “What do you notice about the labels?” Then wait. Long enough that the silence feels uncomfortable. The Try This / Consider This section below has the specifics.
This is not slacking off, though it will absolutely feel like slacking off. The version of teaching that fits a child’s wiring most days is the version that does less of the part that exhausts you and more of the part that actually works.
Try This: The One-Question Errand
For one week, pick one ordinary trip a day (the grocery store, the walk to the car, the dishwasher loading, the dinner prep) and run it as a one-question experiment.
1. Pick the moment before you go. The cereal aisle. The cutting board. The pile of mismatched socks.
2. Ask one question instead of giving one instruction. Not “put your shoes on” but “what shoes work for where we’re going?” Not “the big one goes on the bottom” but “which one do you think goes on the bottom?” Not “this is a half cup” but “how much do you think we need?”
3. Wait. Long enough that the silence feels uncomfortable. Most parents step in before kids can formulate the thought. Try waiting them out.
The point of the question is not the right answer. It is to find out what your kid thinks before you tell them what to think. Most of the time, what they say will surprise you, even when it is wrong. That is the developmental work. No need to correct them directly. Sometimes you might use a tablespoon to measure out 2 cups of flour: no harm no foul.
Make a note at the end of each day. One line. What you asked, what they said, and how hard it was for you not to jump in. The third thing is the interesting one.
CONSIDER THIS
When my kid is figuring something out, what makes me jump in?
The reflex to jump in doesn’t fire because you are a bad teacher. It fires because we’ve been taught that telling is good teaching and making mistakes is Bad. The silence while your kid is figuring something out feels like a void you are supposed to fill. It isn’t. The silence is where their thinking lives. You are not slacking off when you stop talking. You are giving them the space they need.
Want the research behind this post? Read the Dive Deeper: The Grocery Store Is a Classroom — five studies that anchor what we wrote, what they found, and where the evidence has limits.
Every Functional Child Development post comes with a Dive Deeper companion. Subscribe for the full picture.