There is a specific kind of tiredness that comes from the twenty seventh "but why" of a single car journey. It is not the tiredness of a hard day. It is the tiredness of having genuinely tried to explain photosynthesis to someone who is four, and being asked why again.
The reassuring news is that the number is not in your head, and neither is the exhaustion. Researchers who have counted find rates that would break most adults. The more useful news is that decades of studies now show fairly precisely what children are doing when they ask, what kind of answer actually satisfies them, and what happens when they do not get one. Understanding that turns the twenty seventh why from an endurance test into something you can answer efficiently.
How many questions does a preschooler actually ask?
Around 107 an hour, of which roughly 76 are genuine attempts to get information. Those figures come from Michelle Chouinard's monograph Children's Questions: A Mechanism for Cognitive Development, published by the Society for Research in Child Development in 2007, which analysed transcripts of naturalistic parent and child conversation from the CHILDES database. The rest are requests, clarifications and social noise.
Harvard child psychologist Paul Harris has put the cumulative total at roughly 40,000 questions between the ages of two and five. What changes across those years is not the volume so much as the type. Chouinard found that between two years zero months and two years five months, only about 4 percent of a child's questions asked for an explanation. By age five, that had risen to around 30 percent.
Chouinard's core sample was small, four children for the main analysis, which is a real limitation worth stating. But the pattern has held up across later work, including a 2026 review in Child Development Perspectives that pulls together two decades of question asking research.
Why do children ask "why" so much between 3 and 6?
Because somewhere around age three, children work out that other people know things, and that those things can be extracted by asking. It is a genuine cognitive breakthrough rather than a personality trait, and once a child has it, questioning becomes by far the most efficient learning tool available to someone who cannot yet read.
The underlying mechanism is that children notice a gap in what they know and go looking for someone to fill it. Work reviewed by Boston University's Child Development Lab shows children direct their questions strategically, choosing adults who have previously given them useful information. They are not broadcasting. They are selecting.
This is also why the phase overlaps with other big language changes at the same age. A four year old asking constantly is the same four year old rapidly expanding vocabulary, which we mapped out in our piece on vocabulary milestones by age, and narrating their own actions out loud, which we covered in why children talk to themselves. All three are the same engine running.
Do children really want the answer, or just the attention?
They want the answer, and there is evidence that distinguishes the two. Frazier, Gelman and Wellman, publishing in Child Development in 2009, looked at what preschoolers did immediately after asking a causal question, depending on what they got back.
When an adult gave a real explanation, children tended to agree and then ask a follow up question that built on it. When the adult gave a non explanation, children tended to re-ask the original question, or supply their own explanation instead. That asymmetry is the tell. A child fishing for attention would be satisfied by any response. A child who repeats the question is telling you the answer did not land.
Which reframes the twenty seventh why. It is usually not the twenty seventh question. It is often the same question, still open, phrased again because the first three attempts did not resolve it.
What counts as a good answer?
One that names a mechanism rather than restating the fact. Kurkul and Corriveau, in Child Development in 2018, showed that children not only prefer non circular explanations but remember them better afterwards. Related work found that five year olds detect circular explanations and subsequently trust that informant less on new topics.
That last part is the one worth sitting with. A circular answer does not just fail to inform. It quietly downgrades you as a source.
| The question | Circular answer | Mechanism answer |
|---|---|---|
| Why do leaves fall? | Because it is autumn | The tree stops feeding the leaves so it can save energy for winter, and once they stop being fed they drop off |
| Why do I have to sleep? | Because it is bedtime | While you sleep your body repairs itself and your brain files away everything you learned today |
| Why is the sea salty? | Because it has salt in it | Rain washes tiny bits of salt out of rocks, rivers carry them to the sea, and when the water evaporates the salt stays behind |
| Why does the moon follow us? | It just looks that way | It is so far away that moving the car barely changes the angle you see it from, so it seems to stay level with us |
None of these require expertise. They require one extra clause: the bit that says how, not just that.
Does it matter who answers, and how often?
It does, and this is where the research gets uncomfortable. Kurkul and Corriveau recorded thirty seven four year olds in conversation with their caregivers and found that children across socioeconomic groups asked a similar proportion of questions. What differed was the response. Caregivers in the mid socioeconomic group offered significantly more explanatory answers, and significantly more non circular ones, than caregivers in the lower socioeconomic group. The full paper is available from Boston University.
The children's curiosity was equal. The information they got back was not. That is a finding about time, stress and available bandwidth rather than about how much anyone loves their child, and it is worth naming because it points at the actual lever. Nobody can answer 107 questions an hour well. But moving from mostly circular to sometimes explanatory changes what a child takes away.
This is also the honest case for anything that adds another patient answering voice to a child's day, whether that is a grandparent, an older sibling, a library visit or a talking toy. We built Ted to answer children's questions at their level and in their language, and the reason is exactly this gap. It is not a replacement for you, and no evidence suggests it should be. Research on question asking in preschool classrooms shows how much depends on a responsive adult who knows the child. What an extra source can do is absorb some of the twenty seventh why so that the twenty eighth still gets a real answer from you.
What do you say when you do not know, or cannot face another one?
"I don't know, how could we find out?" is a legitimate and well supported answer, because it models the actual process rather than faking the result. Studies of preschoolers' information search strategies show children are already building rough theories about where knowledge lives, and showing them where you look is content in itself.
Four answers that work when you are out of energy
- "What do you think?" Often produces a theory you can then build on, and takes the load off you.
- "I don't know. Let's find out after dinner." Only use it if you will actually come back to it, because children track this.
- "Ask me again in the car." Deferring is fine. Dismissing is what teaches them to stop.
- One clause, then stop. A short mechanism beats a long lecture. You do not need the full explanation, just the how.
Do children stop asking, and does that matter?
Question rates do fall as children move through school, and the reasons are debated rather than settled. Part of it is developmental, since older children can read and search independently. Part of it appears to be environmental: classroom studies consistently find that teachers ask most of the questions and children ask remarkably few.
What the evidence does not support is the popular claim that curiosity is destroyed at a fixed age. The safer reading is that children keep asking where asking works. If a question reliably produces an answer with a mechanism in it, the questions keep coming. If it reliably produces "because I said so", the child updates. That is the same selective trust mechanism Kurkul and Corriveau measured, applied over years instead of minutes.
So the twenty seventh why is not a test of your patience. It is a small ongoing assessment of whether you are worth asking. You will fail some of them, and that is fine. You only need to pass most.
Frequently asked questions
How many questions does a 4 year old ask per day?
Observational research puts the rate at roughly 107 questions per hour of conversation with a parent, about 76 of which seek information. Across the whole of the preschool years, one widely cited estimate is around 40,000 questions between ages two and five.
At what age do children start asking why?
Explanation seeking questions appear from around age two but are rare at first, making up roughly 4 percent of questions in the first half of the third year. They rise steadily and reach around 30 percent by age five, which is why the why phase feels like it arrives at three or four.
Is it normal for a child to ask the same question repeatedly?
Yes, and it usually means the previous answer did not satisfy them. Research shows children re-ask a question when they receive a non explanation, and move on to a follow up question when they receive a real one. Repetition is feedback rather than defiance.
What is a good way to answer a child's why question?
Give a mechanism rather than restating the fact. "Because it is autumn" is circular. "The tree stops feeding the leaves so it can save energy for winter" explains how. Children remember mechanism answers better and trust the person who gives them more on future topics.
Should I worry if my child does not ask many questions?
Not on its own, since children vary widely and some ask more through action than speech. It is worth mentioning to a health visitor or doctor if it comes alongside limited vocabulary, little interest in back and forth conversation, or a loss of skills the child previously had.