The Cognitive Skill That Predicts Everything, and the Activities That Build It

A child who can hold three instructions in her head, resist blurting out the first answer that comes to mind, and switch strategies when the first one isn't working will outperform a child who can't do those things. Not in one subject. In all of them. And not just academically: in friendships, in sport, in every situation that requires thinking before acting.

That cluster of abilities has a name. Researchers call it executive function. And a growing body of controlled studies, across robotics, coding, structured play, physical challenges, and science experiments, is converging on a finding that matters for parents: executive function in young children can be trained, the training doesn't have to look like work, and the gains show up in domains that have nothing to do with how they were built.

What is executive function and why it matters?

Executive function is the umbrella term for three cognitive processes that develop rapidly between the ages of three and seven, supported structurally by the prefrontal cortex.

Working memory is the ability to hold and manipulate information: remembering a set of directions while walking across a room, or keeping track of a multi-step maths problem without losing the thread. Inhibition is the ability to override a reflexive response: not grabbing, not blurting, waiting your turn, choosing the considered action over the automatic one. Cognitive flexibility is the ability to shift between tasks or perspectives: adjusting when the rules change, seeing a problem from a new angle when the obvious approach has failed.

These are not academic skills. They are the infrastructure underneath academic skills. A child with strong executive function can follow a multi-part instruction, recover when something doesn't go to plan, and stay focused when something more interesting walks past the window. A child with weaker executive function struggles with all three, and no amount of subject knowledge compensates.

A landmark longitudinal study tracked children's self-control (a close proxy for executive function) at ages 3-11 and followed them for 30 years. Those with better self-control in childhood had better health, earned more, and committed fewer crimes as adults, even after controlling for IQ, gender, and social class. The researchers concluded that even small improvements in self-control in childhood could shift outcomes at a population level. Executive function is not a nice feature of cognitive development. It is foundational architecture.

The robotics evidence: 187 children, measurable gains

The largest and most rigorous study in educational robotics and executive function is a randomised controlled trial from the University of Pisa, published in Frontiers in Psychology. A team from IRCCS Fondazione Stella Maris enrolled 187 typically developing five- and six-year-olds and randomly assigned them to either ten weeks of robotics programming or a waitlist control.

The intervention was simple. Children used a Bee-Bot, a palm-sized floor robot that moves in fixed steps when buttons on its back are pressed. No screen. The child plans a sequence of moves, presses go, watches the result, and debugs when it goes wrong. Sessions ran twice a week, in small groups, with activities that became progressively harder over 20 sessions.

After ten weeks, the robotics group showed statistically significant improvements on standardised neuropsychological tests of visuospatial working memory and inhibition. Not robotics tests. Clinical instruments that measure general cognitive function. The control group, continuing with their regular school curriculum, showed no equivalent gains.

The mechanism maps cleanly onto the demands of the task. To program a floor robot, a child must hold a goal in mind, plan a sequence, predict the outcome before executing, and then compare the result to the intention. If it doesn't match, they identify the error and revise. That cycle is a structured workout for working memory and inhibition.

A separate study tracking 6-8 year olds with eye-tracking technology over six months of robotics found visuospatial working memory improving at roughly 4% every two months. And research on coding in Italian first graders found that even one month of block-based programming improved planning and response inhibition on unrelated tests. The Pisa RCT adds the rigour the earlier studies lacked: large sample, random allocation, and a waitlist control.

What the non-robotics research says

Here is where the picture gets more interesting than a story about robots.

In 2011, Adele Diamond published a review in Science examining every approach that had been shown, in controlled studies, to improve executive function in children aged 4 to 12. The list was broader than most parents would expect: computerised working memory training, non-computerised games, aerobic exercise, martial arts, yoga, mindfulness, Montessori education, and specific school curricula like Tools of the Mind.

What linked them was not content. It was structure. The approaches that worked all shared the same features: repeated practice with progressive difficulty, genuine cognitive challenge (not passive participation), and activities that required the child to hold information in mind, resist impulse, and adjust their approach. The domain was irrelevant. The cognitive demand was everything.

A 2025 randomised controlled trial with 62 children aged 4-5 tested three types of scaffolded play, each delivered over ten weeks: movement play, puzzle play, and language play. All three groups improved significantly in inhibition and cognitive flexibility compared to the control group. The gains were sustained at a three-month follow-up. Language play was the most consistent across all three executive function components, but movement and puzzles both produced meaningful improvements.

A separate 2025 RCT with 80 children aged 4-6 found that 12 weeks of structured motor learning (30-minute sessions twice a week) improved working memory compared to outdoor free play. These were not robotics sessions or coding lessons. They were physical challenges designed with progressive difficulty and clear goals.

The 351-child randomised controlled trial of Tools of the Mind, a play-based kindergarten curriculum at the University of British Columbia, found that children in classrooms emphasising hands-on learning, collaborative play, and structured challenges showed better executive function and academic outcomes than those in traditional teacher-led classrooms, even when the traditional classrooms spent more time on academic content.

The pattern across these studies is consistent. What builds executive function is not any particular subject or activity. It is any activity where a child has to plan, hold information in mind, resist the urge to act before thinking, adjust when something doesn't work, and do all of that with tasks that get gradually harder. Robotics does this. So does building a bridge from popsicle sticks, running an obstacle course that changes every round, or following a multi-step science experiment where the outcome is not guaranteed.

Why the early years matter most

Executive function develops rapidly in the preschool and early primary years, supported by structural changes in the prefrontal cortex. This is not a claim that the window closes. Executive function continues to develop into adolescence and responds to training across the lifespan. But the rate of development is highest between roughly three and seven, and the architecture built during that window provides the foundation for everything that follows.

A longitudinal study from the University of California, Irvine tracked 1,096 children from birth using data from the NICHD Study of Early Child Care. They found that the quality of cognitive stimulation children received in early care settings predicted their STEM achievement in Grades 3-5, which in turn predicted STEM achievement and school performance at age 15. The pathway was indirect but clear: early stimulation built foundational skills that made later learning easier, and those advantages compounded over a decade. The effect was strongest for children from lower-income backgrounds.

This doesn't mean you need to enrol your five-year-old in a neuroscience-approved training programme. It means that the things many parents are already doing, or could start doing this week, are building cognitive architecture that will matter for years. The key is structure: not rigid instruction, but activities that present a genuine challenge, require thinking before acting, and get progressively harder as the child's capacity grows.

What this doesn't prove

Executive function is influenced by many things: sleep, nutrition, emotional regulation, the quality of relationships, genetics. No single activity, however well designed, operates in isolation.

The robotics RCT involved 187 children, which is large for its field but small by medical standards. The intervention was delivered by trained researchers, not parents at a kitchen table. The play and movement studies had smaller samples and shorter follow-ups.

Most of the controlled studies involve children aged 4-7. Whether the same interventions produce equivalent gains in older children, children with learning differences, or children in different cultural contexts is less well established.

And the gains, while statistically significant, were moderate. These activities do not transform cognitive function overnight. They build it incrementally, the same way physical exercise builds cardiovascular fitness: through repeated effort, progressive challenge, and consistency over time.

What this means at home

You do not need a robot. You do not need a coding platform. You do not need any specific product or programme.

What the research consistently shows is that executive function improves when children engage in activities with a specific structure: a clear goal, a sequence of steps to get there, enough difficulty to require genuine thought, feedback that shows whether the plan worked, and the opportunity to revise and try again. The content can be anything. The structure is what matters.

A child building a popsicle-stick catapult is planning, predicting, testing, and adjusting. A child following a recipe with three steps she has to hold in her head is exercising working memory. A child playing a physical game where the rules change every round is practising cognitive flexibility. A child programming a robot to cross a mat is doing all of it at once.

None of it feels like training. That is rather the point.

Research sources

Di Lieto, M. C., Inguaggiato, E., Castro, E., Cecchi, F., Cioni, G., Dell'Omo, M., Laschi, C., Pecini, C., Santerini, G., Sgandurra, G., & Dario, P. (2020). Empowering executive functions in 5- and 6-year-old typically developing children through educational robotics: An RCT study. Frontiers in Psychology, 10, 3084. Randomised controlled trial, 187 children. University of Pisa / IRCCS Fondazione Stella Maris. https://pmc.ncbi.nlm.nih.gov/articles/PMC7012808/

Diamond, A. (2011). Activities and programs that improve children's executive functions. Science, 333(6045), 959-964. Review of interventions shown to improve EF in children aged 4-12. University of British Columbia. https://www.science.org/doi/10.1126/science.1204529

Moffitt, T. E., et al. (2011). A gradient of childhood self-control predicts health, wealth, and public safety. Proceedings of the National Academy of Sciences, 108(7), 2693-2698. Longitudinal study, 1,000 children followed 30 years. https://doi.org/10.1073/pnas.1010076108

Bustamante, A. S., Bermudez, V. N., Ochoa, K. D., Belgrave, A. B., & Vandell, D. L. (2023). Quality of early childcare and education predicts high school STEM achievement for students from low-income backgrounds. Developmental Psychology, 59(8), 1440-1451. Longitudinal study, n = 1,096. University of California, Irvine. https://pmc.ncbi.nlm.nih.gov/articles/PMC10524717/

Zhou, Y., et al. (2025). The effect of child-appropriate play with various types on preschoolers' executive function. Acta Psychologica. Randomised controlled trial, 62 children aged 4-5, 10-week intervention with 3-month follow-up. https://doi.org/10.1016/j.actpsy.2025.104711

Diamond, A., & Ling, D. S. (2019). Fundamental questions surrounding efforts to improve executive functions (including working memory). In Cambridge Handbook of Cognition and Education. University of British Columbia.

Diamond, A., et al. (2019). Randomized control trial of Tools of the Mind: Marked benefits to kindergarten children and their teachers. PLOS ONE, 14(9), e0222447. RCT, 351 children, 18 schools. https://doi.org/10.1371/journal.pone.0222447