Guide · Persistence · last reviewed 20 September 2026
The Child Who Gives Up Immediately
A child who pushes the thing away after one go is telling you what the failure cost, not what they are made of. Whether anybody tries again, child or adult, depends far more on how expensive the mistake was than on how determined they are. A peg that falls off the rim of a bowl can be picked straight back up. A tower of forty blocks that collapses cannot. Four changes do almost all of the work, and every one of them is to the material rather than to the child: make the failure cheap, make the error visible without an adult in the room, put out a spare, and wait about ten seconds longer than feels comfortable before helping.
This is a different problem from waiting
Parents use one word, patience, for three things that behave completely differently, and the guide on how patience actually develops separates them: waiting for something, tolerating a slow process, and persisting through difficulty when the lace will not go through the hole.
That third one is this page, and it is not really about time at all. It is about frustration. A child who cannot wait four minutes for a biscuit may sit and thread the same bead eleven times, and a child who will happily watch a bean for a fortnight may throw the bolt across the room after two turns. Those are not contradictions and neither of them is a measure of character. They are two different tasks with two different costs of failure.
The cost of the mistake decides the second attempt
Look at what is actually lost when it goes wrong. On Pattern Card Bead Lacing, one bead falls off the lace, not the whole string. On Cardboard Loom Yarn Weaving, one row is wrong, not the cloth. On Clothespin Pinching the peg lands on the table and the hand picks it up.
Now take the forty-block tower. Twenty minutes of work, and one error costs all of it. A child who stops trying at that is doing the arithmetic correctly: work that punishes a mistake by destroying everything before it teaches a sensible person to stop, fast.
This is why the persistence shelf looks the way it does. Posting, pegs, slots, pinching, sponges: twenty activities where a failed attempt costs a disc on the table and the pile is still sitting there. They are not easy. They are cheap to get wrong, and cheapness is what buys the eleventh try.
Where the cost genuinely is high, the move is to make the recovery cheap instead. Soap snaps, usually across the narrow direction of the bar. On Soap Carving and Safe Whittling the answer is not a fresh start, it is carrying on with the larger half, which is easier to hold anyway. Nobody has to begin again, and the next shape gets drawn wide rather than tall.
The error has to be visible without you
On Peg and Hole Matching the peg either stands up or it does not. There is no adult in that loop. The board answers on its own, immediately, in a way a two year old can read, and that is what makes the second attempt the child's rather than a response to something you said.
When only an adult can tell whether it worked, every attempt needs that adult and the run is over the moment you leave the room. So it is worth checking before you set an activity out: can the child see the result without asking? A jammed nut is obvious and so is one that runs freely. The coins fall or they stay. The disc goes through or it sits on the lid.
The same mechanism is why it is worth letting a material make a point that you would otherwise have to make. On Balance Beam Coin Carry, add coins one at a time until the stack falls on a fast walk and survives a slow one. Most children find the right speed themselves inside three or four crossings, and nobody has been told to slow down.
Visible does not mean loud, though. Do not count the drops out loud and do not narrate the near misses, because a running commentary turns a self-correcting activity back into an assessment with an examiner in the room.
Put out one more than is needed
This is the smallest change on the page and it has the largest effect. Two bars of soap on the table rather than one turns a snapped carving from a disaster into a shrug. Thirty pegs instead of ten. Six pattern cards in a stack rather than the one being worked on. A second sheet of paper sitting beside the one being folded.
The spare works before anything goes wrong, which is the part that is easy to miss. A child who can see that there is another one takes bigger risks with the first, and taking a risk with the first is where most of the learning is. On the soap, the second bar has an extra job: handing over a spare bar to hack at with no shape in mind at all is usually what gets a six year old past the point where it stopped feeling like it was working.
A pile also sets the length of a sitting. Eight discs is eight posts; on Cardboard Slotting and Coin Drop, thirty discs cut round a mug roughly triples it. Nine holes is nine decisions, and at two that can be the whole activity, so a bigger board is the change that works rather than a longer instruction.
Wait about ten seconds longer than feels comfortable
The unhelpful move is the natural one, and everyone makes it: stepping in at the exact moment of difficulty. Taking the lace and threading it yourself ends the activity and takes the result with it, because the difficulty was the activity. The thread that would not go through was not in the way of the work. It was the work.
What we see is that an adult's hand arrives at around three seconds, and a child's own solution somewhere between four and nine. That means the help usually lands in the gap, just before the thing was about to happen. Trying a slow count of ten, once, is the whole intervention. It feels much longer from the outside than it does from the inside.
When a child does say what went wrong, repeat it back without adding a solution. Very often they solve it themselves in the next thirty seconds, and the sentence they said is the more valuable half anyway, because it means the error has become theirs to work on.
When a child asks directly for the hard part to be done, alternating beats both refusing and taking over. Say you will do one cut and she does the next, then hold to that exactly. Two rounds is usually enough for the tool to come back across the table.
What giving up is telling you about the level
Every activity page in this library has a section for what to do when it goes sideways, and they are all written the same way round: the thing that went wrong is information about the level, followed by the one change that answers it. It is worth reading a few of them together, because the pattern is the point.
The nut stops dead after one turn. It went on crooked and jammed, and this is the main reason a three year old walks away from Nuts and Bolts Screwing. Take it right off, sit it flat, and turn it backwards until it drops level. Showing that backwards half turn once is the most useful thing there is to teach on that page.
He gives up at the fourth step of the paper cup. Step four is where the model has stopped looking like a square and does not yet look like a cup, so there is nothing to keep going for. On Paper Folding Precision Challenge, do the last two steps together for the first three attempts. Once he knows what it turns into, the middle stops being a cliff.
He stopped halfway through the weaving and has not been back. The middle is the slow part: the rhythm has lost its novelty and the end is not in sight. Change to a much thicker yarn, because a chunky row fills four times the height of a thin one and visible progress is what brings a child back.
She cannot get the peg in at all. The hole is too tight and wants adult force rather than two year old force. One turn of the pencil opens it a millimetre, and the peg should then drop in under its own weight.
Not one of those is a fact about the child. Every one is a fact about a slot, a thread, a step or a piece of yarn, and every one has a fix that takes under a minute. That is the honest limit of what an activity can be said to do, too. We are not going to tell you that a tray of tweezers builds grit, and the field's own instruments do not support that kind of precision either: as Diamond puts it, almost no executive function measure requires only one executive function.
What a well-pitched activity reliably does is make the next attempt cheap enough to be worth making.
Diamond, A. (2020). Executive functions. Handbook of Clinical Neurology, 173, 225 to 240. doi:10.1016/b978-0-444-64150-2.00020-4
What this looks like at two, at four and at six
At two and three, expect three or four tries and then the object gets pushed away, or thrown. Both are information about the level rather than about temperament. The move is to make the target bigger: a wider slot, a fatter lace, a softer peg, a bigger hole. One step easier and the trying restarts, usually straight away. At this age the length of a run also follows the size of the pile more closely than anything else you can change.
At four and five, eight or ten tries is common, and the new thing is that the child will say what went wrong. That sentence is what to listen for. It means the error has moved inside, and from there the useful adult move is to say it back and then be quiet.
At six and seven, a child will undo good work to fix a mistake nobody else noticed: two rows of weaving pulled out over a hole at the edge, a whole side of a carving taken back. Say yes to that even when it costs the afternoon. The choice to redo it is the whole thing.
What is worth doing this week
Put out two of whatever it is. Two bars of soap, two looms, thirty pegs instead of ten. It is the smallest change here and the one that shifts the most.
Try a count of ten before helping, once. Just once, to find out where in that gap your child's own solution lives.
Check one activity for whether it answers its own question. If only you can tell whether it worked, either change the material or plan to stay, and choose which on purpose.
Pick the rung rather than the age. Nearly every page here has four rungs on one set of materials, and dropping back a rung for a fortnight is the ordinary way through a wall. There is more in the guides on making something harder without making it hard and when it does not work.
The shelf itself is twenty activities from two to seven, sorted so the cheap failures sit at the start of it.