Training One Skill Moved a Neighbouring One

What the research found

Executive function is the family of mental skills that lets a child hold a goal in mind, resist distraction, and switch flexibly between tasks. For years, the hope has been that you could strengthen one of these skills through repeated practice and watch the benefit spread to others. That spreading benefit has a name: far transfer. It has also been the hardest thing in this field to actually demonstrate.

This 2022 randomised controlled trial, run by Lauren Irwin Harper and colleagues at Florida State University, tested the idea directly. Fifty-four children with ADHD, aged eight to twelve, completed a ten-week training programme. They were randomly assigned to one of two arms. One group did central executive training, which targets working memory—the ability to hold and manipulate information in mind. The other did inhibitory control training, which targets the capacity to stop an automatic response. Each child had a weekly monitored session and practised at home. Crucially, neither programme trained set shifting, the skill of switching between rules or tasks. Set shifting was the outcome the researchers measured—the untrained skill they hoped might improve.

The two programmes did not perform equally. Children who trained inhibitory control improved at set shifting; children who trained working memory did not. On the primary measure of set-shifting accuracy, the difference between the groups favoured inhibitory-control training with an effect size of d=0.63. Broken down by task, the inhibitory-control group improved by d=0.81 on one set-shifting measure and d=0.51 on another. The working-memory group showed negligible change on both, with effect sizes near d=0.10. A parallel pattern appeared for the speed of switching: the inhibitory-control group improved (d=0.59) while the working-memory group did not.

An effect size of d is a way of expressing how large a change is relative to the spread of scores. Values around 0.5 are conventionally called medium, and values approaching 0.8 large. So the headline is that training inhibitory control produced medium-to-large gains on a skill it never directly practised, whereas training working memory produced essentially nothing on that same skill. The authors also reported Bayesian evidence—a measure of how strongly the data favour one conclusion over another—that supported the inhibitory-control advantage rather than chance.

It is worth being precise about what this does and does not show. It is a genuine randomised trial, which means the comparison between the two groups is causally interpretable: the training, not some pre-existing difference, is the most plausible explanation for the gap. But it is a single study of 54 children. It measured outcomes immediately after training ended, so we do not know whether the gains last a week, a term, or a year. And the children were selected for below-average working memory, not for weakness in set shifting, so the sample was not built around the skill that ended up improving.

What it means for you and your child

If your child has ADHD and you are weighing a brain-training programme, the honest summary is this: one carefully run trial found that a specific kind of training moved a specific untrained skill. That is more than most of this literature can claim. It is not a promise that the same result will hold for your child, in your home, on the outcomes you care about most.

Notice what the study did not measure. It looked at set shifting on laboratory tasks. It did not report whether children became easier to manage at homework time, calmer in the classroom, or better at the daily transitions—from screen to dinner, from one subject to the next—that actually shape family life. Improvement on a switching task is a real finding, but it is a stepping stone, not the destination. The distance between a laboratory gain and a visible change in behaviour is exactly the distance that has tripped up brain-training claims for two decades.

The most useful signal here is directional rather than prescriptive. The result suggests that inhibitory control—the stop-and-check skill—may be more central to flexible thinking than working memory is, at least for these children. That is a plausible and interesting idea. It is not yet a settled fact. A single study, however clean, is a data point. Confidence comes from replication: other teams, other samples, and follow-up measurement showing the gains survive past the final session. The authors themselves flag these gaps, which is a mark of careful work rather than a weakness in it.

For a practical decision today, treat this as one input, not a verdict. Ask any programme you are considering what outcomes it has actually improved, in whom, and whether those gains were still there months later. Be sceptical of any product that borrows the language of a study like this while quietly promising far more than the study delivered. The researchers demonstrated a modest, specific transfer effect under controlled conditions—not a general upgrade to a child’s thinking.

None of this diminishes the value of the ordinary, well-evidenced supports: structure, clear routines, help with transitions, and where appropriate the treatments your clinician recommends. If a training programme appeals to you, approach it as a reasonable experiment with realistic expectations, not as a shortcut. The most encouraging thing about this trial is not any single number. It is that the field is now capable of asking the transfer question rigorously and answering it honestly—including, when the evidence points that way, saying that one approach worked and another did not.