A note on scale: this piece begins with two children on two trails at around the same time, and with a habit of route design we have kept for years. One study appears later. We found it after the habit, not before.
Late May, on Taiwan’s northeast coast. A group of eight- and nine-year-olds had come, for the first time, to the place where the mountains meet the sea.
The route that day crossed a stretch of bare rock. One child crouched down there — ran a hand over the grain of the stone, then followed the water lines the retreating waves had left on its surface, watching them slowly disappear.
He crouched there for a long time.
The leader saw him and did not walk over. Did not ask what he was doing, did not remind him the group was moving on, did not take the chance to explain what kind of rock it was.
Around the same time, a kindergarten group was walking a trail in New Taipei. The notes from that day contain one line:
Some children walked fast the whole way; some stopped for a very long time to look at a butterfly.
Both look like gaps in the itinerary. They are not.
Leaving space is not taking structure away. It is leaving, inside the structure, a stretch of time with no assigned purpose.
This piece is about how that sentence gets carried out in a single day’s itinerary.
Those ten minutes were on the itinerary
When we design a route for six- and seven-year-olds, we deliberately do not push the difficulty up. The reason is not consideration for their comfort. For the kind of day we are designing at this age, we want the child to have something left in reserve. When most of a child’s effort is going into simply finishing the route, there may be less room left for noticing what is around them: the bark of a tree, a squirrel in the branches.
So one of the conditions of the route is “plenty of stopping points”. A stopping point is not where you stop because you cannot go on. It is scheduled in advance.
The same logic extends to what adults do. The leader’s working principles on the trail are: don’t hurry them, don’t over-protect them, and don’t over-explain. That last one is the easiest to forget. With a child crouched in front of a rock, all an adult has to do is walk over and ask “What are you looking at?”, and those ten minutes are over.
We have done this for many years, and the reason has always been experiential.
We began to notice, again and again: when a child is observing, handling or simply drifting on his own, and an adult quickly takes over with questions and explanations, we have sometimes seen the child’s attention shift from what was in front of him towards the adult and whatever comes next.
That is a field observation, not a measurement. We had no comparison group, kept no records, and cannot rule out that the child was simply tired that day.
It was only later that we found someone had studied a question nearby.
A study about six-year-olds
Barker and colleagues, in a study published in Frontiers in Psychology in 2014, worked with 67 children with an average age of six and a half.
They asked parents to report how the child’s previous week had been spent, in thirty-minute blocks, sorted into two kinds: “structured” (adult-led lessons, homework, extracurricular classes) and “less structured” (free play, reading, family outings).
They then measured one ability, which the authors call self-directed executive function. The measure was a verbal fluency task: the child has to name words from one broad category within a time limit, and the score is not how many words come out, but how the child decides, on his own, when to switch from one semantic sub-category to the next.
The result: more time in less-structured activities was associated with better performance on this task.
And one detail matters a great deal. The relationship appeared only on the self-directed measure. It did not appear on executive-function tasks driven by external instruction.
Put more precisely: the difference showed up in the ability that requires a child to decide for himself what to do next and when to switch, not in the ability to complete a task according to someone else’s instruction.
That is close to what we keep noticing in the field: when an adult holds off from giving the next instruction, some children immediately start finding something of their own to do, and some stay where they are and wait.
“Close to” is not “the same as”. Switching categories in a verbal fluency task and deciding whether to crouch down on a trail are two different kinds of observation. The study made our judgement more precise. It did not prove what we see in the field.
What this study does not say
This has to be said plainly: it is a correlation, not a cause.
The authors themselves list three possibilities they could not rule out: perhaps children who already have stronger self-directed abilities choose more less-structured activity; perhaps the causation runs the other way; perhaps some unmeasured factor, such as the parents’ own executive function, affects both sides at once.
The sample was 67 children, all in the United States, average age six and a half.
And one detail that matters for parents in Taiwan: the study counted reading and media use as less-structured time. So it cannot be read as “letting a child zone out in front of a screen works just as well”; the study did not test that. It is worth noting, though, that the relationship still held after the researchers excluded screen and media time, so the overall result is not being carried by screen time.
We did not design our routes this way because we saw this study. The order was the reverse: we noticed the same thing in the field over and over, and only later found that someone was studying a nearby question.
Leaving space is not leaving children to it
This is the point most easily misread, so it needs stating.
What that child could not see was this: the risks on the route had been assessed in advance, and the group’s position, communication and the pace of gathering everyone up were being kept in hand throughout. He could, for a while, stop dealing with “will I get left behind?”, and that is what left him the room to give his attention to the rock in front of him.
The purpose of designing the surrounding structure is precisely so that children do not have to carry the group’s safety and pace themselves, and so keep more attention for what is happening in front of them.
So the sentence bears repeating, because it dismantles two opposite extremes:
Leaving space is not taking structure away. It is leaving, inside the structure, a stretch of time with no assigned purpose.
At one extreme, “the fuller the schedule, the more educational value”. At the other, “only leaving children entirely to their own play counts as autonomy”. What we do is a third thing: adults design the conditions. They don’t decide what has to happen inside them.
How we leave space: three kinds of margin
In designing a day outdoors, we treat that space as three kinds of margin: margin in difficulty, margin in time, and margin in how much adults step in. They describe how we plan; they are not steps to follow.
Margin in difficulty. When choosing a route, we make “the body still has something left at the end” a condition, rather than pushing the difficulty to the upper limit of what a child can just about bear. A high-challenge route can bring a different and important kind of experience, but we cannot assume that a child whose physical resources are heavily taken up will still have the same room to perceive. The designer has to know whether this section is going after challenge or after space, rather than cramming both values to the limit.
Margin in time. Stopping points are written into the itinerary, not squeezed out at the end once everything else has been compressed. From a design point of view, a scheduled stop and a stop that exists only because “we happened to walk faster, so there are ten spare minutes” are different things. The first is protected in advance; the second can disappear as soon as the schedule tightens.
Margin in how much adults step in. Don’t over-explain. When a child crouches down, don’t ask what he is doing, and don’t use the moment to add some knowledge. An adult’s silence is also a kind of space — and the hardest kind to actually do, because it looks so much like doing nothing.
Bringing this judgement home
At home, the closest equivalent is not setting aside a prescribed number of minutes of “free time”. It is something smaller:
When a child has stopped at something of his own accord, don’t be in a hurry to turn that moment into a conversation, a lesson or a task.
Sometimes you can wait a little first.
Those few minutes may produce nothing. They never needed to.
An honest limitation
We have not measured anything. The three kinds of margin above are drawn from fifteen years of practice in the field; they are not experimental results. We have not run a control group, and we have not followed up on any of the children who crouched in front of a rock.
The Barker study is correlational, with 67 children, in the United States, average age six and a half. It cannot be applied directly to eight- and nine-year-olds in Taiwan. And it measured a laboratory task, not behaviour on a trail.
We are not aware of any systematic survey of any scale on children’s unstructured time in Taiwan. That gap is worth pointing out.
Nor do we know how much space is enough. To this day we judge it on the ground, without a standard.
Finally
That child eventually stood up, caught up with the group and kept walking.
We do not know what those ten minutes left behind, and we did not ask him to say.
Nobody told him what to look at. No question was waiting for an answer.
In those ten minutes, what to attend to, how long to look, and when to leave were all his to decide.
That is why we left those ten minutes on the itinerary.