Executive Function Becomes Itself at Two

What the research found

Executive function is the family of skills that lets a child hold a goal in mind, resist a distracting impulse, and shift attention when the situation changes. It is what a five-year-old uses to wait her turn, to keep the rule “point to the opposite colour” in mind, and to stop reaching for the toy she was told to leave alone. This study asked a simple question about it: does the executive function we can measure in a baby tell us anything about the executive function that same child will have years later?

The design was unusually patient. Broomell and Bell followed 410 children (201 boys, 209 girls) from the mid-Atlantic United States across six visits: at 5 months, 10 months, 24 months, 36 months, 48 months, and again at 9 years of age. The data were gathered over fifteen years, from 2002 to 2017. At each age the children did the tasks appropriate to that stage, from the classic A-not-B reaching game in infancy, through Simon Says and day/night rules in the toddler and preschool years, to the Wisconsin Card Sort and an N-back memory task at nine.

The central finding is about when the skill starts to stay put. In infancy, it does not. The statistical path from executive function at 5 months to 10 months, and from 10 months to 24 months, was not significant. A baby who did comparatively well on the infant task was no more likely than any other to do well a year later. Then, around the second birthday, something changes. Performance at 24 months predicted performance at 36 months strongly (standardised coefficient 0.90), 36-month scores predicted 48-month scores (0.63), and 48-month scores predicted performance all the way out at 9 years (0.62). Each of these paths was statistically reliable; the infant ones were not.

The authors treated executive function as a single underlying ability rather than several separate ones, and a model built that way fit the data well (RMSEA 0.01, CFI 0.94). Read plainly, the picture is this. Executive function in the first two years is real but not yet stable: it does not travel with the child in a way we can track. From toddlerhood onward it becomes an enduring individual trait, and where a child sits relative to peers at two or four gives a meaningful, though far from complete, indication of where they will sit at nine.

It is worth being precise about the size of that indication. A coefficient of 0.62 is a strong developmental signal, but it is not destiny. It leaves a great deal of room for individual children to move up or down. This is a correlational study of natural development. It shows that early and later executive function are linked; it does not show that the early state causes the later one, nor that intervening at two would change the nine-year outcome. That is a different question, and this design cannot answer it.

What it means for you and your child

The most useful takeaway is also the most calming: do not read too much into a toddler’s, and especially an infant’s, apparent self-control. In the first two years, individual differences in these skills simply do not predict the later child. A baby who seems slow to inhibit a reach, or a one-year-old who cannot yet wait, is not showing you a fixed feature of who they will become. The study found no thread connecting infant performance to anything downstream. Worry here is misplaced.

From around two, the account shifts. Differences between children begin to persist. This is not a warning; it is a clarification of timing. It tells you that the toddler and preschool years are when executive function starts behaving like a stable characteristic, which is a reason to take the ordinary supports of that period seriously: predictable routines, games that ask a child to hold and follow a rule, and adults who model waiting and switching rather than demanding them. The study does not test whether these things help. But it does identify the window in which the trait is forming and firming, which is the window worth attending to.

Be careful, though, about what a strong coefficient means for one child. The link from age four to age nine is robust across 410 children, yet it describes a tendency across a group, not a prediction for an individual. Plenty of children in a sample like this will end up higher or lower than their preschool standing suggested. A four-year-old who struggles with impulse control has not been assigned a nine-year-old self. The correlation is real and the scatter around it is wide, both at once.

Two limits deserve a parent’s attention before drawing conclusions. First, the sample was a healthy community group that was predominantly White and unusually well educated, with 62% of mothers holding university degrees. Findings from such a group may not carry over to children in very different circumstances. Second, the tasks changed with age, and the nine-year measures were done on a computer rather than face to face. Some of the continuity, and some of the gaps, may reflect the tools as much as the child. The authors say as much themselves.

So what should you actually do? Less than the numbers might tempt you to. Support the everyday building blocks of attention and self-regulation in the preschool years, because that is when the skill is taking on its lasting form. Resist both complacency and alarm about an infant’s self-control, because at that age it predicts nothing measurable. And hold the whole picture lightly. This is one careful study of how a trait unfolds on its own. It maps the terrain; it does not tell you that steering the course early will change where a particular child arrives.