Executive Function: The Three Core Skills, and How They Differ From IQ

The three core executive functions

The framework nearly all current research uses comes from a study published in 2000 by Akira Miyake and colleagues. They gave 137 college students a battery of simple tasks, each designed to tap one suspected executive function, and used a latent variable analysis to ask whether those functions were really distinct. The answer was that the three are moderately correlated with one another, but clearly separable — the finding the paper's title calls the unity and diversity of executive functions.

Adele Diamond's 2013 review in the Annual Review of Psychology uses the same three and makes the relationship to everything else explicit: “From these, higher-order EFs are built such as reasoning, problem solving, and planning.” Planning is not a fourth core function sitting beside the other three. It is something the three do together.

Miyake's study also showed that the classic clinical tests of “frontal lobe function” do not each measure the same thing. The Wisconsin Card Sorting Test loaded most strongly on shifting, the Tower of Hanoi on inhibition, and operation span on updating — while dual-task performance was related to none of the three. A test labelled “executive” is measuring a particular part of the system, not the whole of it.

How many executive functions are there? Three, four, seven or twelve

This is the question where the popular sources and the research literature quietly disagree, and it is worth being direct about it because the disagreement is visible inside Google itself.

Search what is executive function in English and Google's “people also ask” box asks “What are the 7 executive function skills?” Run the same search in France and the equivalent box asks “Quelles sont les 4 fonctions exécutives ?” — four. Run it in Greece and it asks “What are the 12 executive function skills?” — twelve. Three markets, three different numbers, none of which comes from the research literature, which settled on three separable components a quarter of a century ago and has kept them.

The longer lists are not wrong so much as differently scoped. They enumerate observable everyday skills — planning, organisation, task initiation, time management, emotional control, sustained attention, goal-directed persistence — which is exactly what a teacher, a clinician or a parent needs to talk about. The three-component model enumerates the underlying processes that statistical analysis can actually separate. A list of twelve skills and a model of three components are answers to different questions, and a reader who has met only the first has no way of knowing that.

If you want one rule of thumb: the number of executive functions a source names tells you what that source is for. Three means it is describing the research model. Seven or twelve means it is describing behaviour.

Is executive function the same as intelligence?

No — and this is the question an intelligence site should answer properly, because the honest answer is more interesting than either “yes” or “no”.

In 2006 Naomi Friedman and colleagues, working with the same three-factor model, tested it directly against intelligence measures: fluid intelligence, crystallised intelligence and Wechsler Adult Intelligence Scale IQ. The paper is titled “Not all executive functions are related to intelligence”, and the result is as clean as the title:

• Updating working memory was highly correlated with the intelligence measures, and stayed strongly related to them in structural equation models that controlled for the correlations between the three executive functions.
• Inhibiting prepotent responses was not.
• Shifting mental sets was not. In the same models, the relations of inhibiting and shifting to intelligence were small and not significant.

The authors' own conclusion is the sentence worth carrying away: current intelligence measures “do not equally assess a wide range of executive control abilities likely required for many ‘intelligent’ behaviors”. In plain terms — an IQ test leans heavily on the one executive function that is about holding and manipulating information, and barely touches the two that are about self-control and flexibility.

That asymmetry explains a great deal of ordinary experience. It is why someone can be genuinely quick and still miss deadlines, and why a high score on a reasoning test does not predict who will keep their temper or change their mind. The same gap shows up in critical thinking, where ability and the habit of using it turn out to be partly independent, and in why cognitive biases affect able people about as much as anyone else.

One more finding is worth knowing, because it cuts against the assumption that executive function is simply a product of upbringing. A twin study by the same group in 2008 found that the common factor shared by the three executive functions was 99% heritable — and, importantly, that this factor “goes beyond general intelligence or perceptual speed”. It is not merely IQ measured a second way. How much of intelligence itself is inherited is covered separately in is intelligence genetic.

Do gifted people have executive dysfunction?

Google's question box asks this, and no page currently ranking for the topic answers it. The research above supplies the reason it is even possible.

If inhibition and cognitive flexibility are largely unrelated to measured intelligence, then a high IQ score carries almost no information about them. Someone can score in the top few per cent on a reasoning test and still find it hard to start a task, hold back an impulse or abandon a plan — not as a contradiction, but because the test never measured those things. The two can vary independently because the research says they largely do.

This is different from the specific, identified profile of a person who is both intellectually gifted and has a diagnosed learning or attention difference, which is covered in twice-exceptional. The point here is narrower and applies to everyone: a reasoning score is not a report on self-regulation, and nobody should read it as one.

When does executive function develop?

Gradually, and over far longer than most single-sentence answers suggest. John Best and Patricia Miller's review in Child Development deliberately looked at studies spanning large age ranges, rather than the preschool samples most reviews concentrate on. Their finding is that executive function “emerges during the first few years of life” and “continues to strengthen significantly throughout childhood and adolescence”, with the three components following somewhat different trajectories.

Diamond's review adds the early detail: inhibitory control is disproportionately difficult for young children, and cognitive flexibility “builds on the other two and comes in much later in development” — children usually cannot pass the standard card-sorting switch task until four and a half to five years old.

You will often see a specific finishing age quoted, usually twenty-five. That is a much more precise claim than the developmental literature makes. What the reviews describe is a long, uneven strengthening with no single endpoint, and different components arriving at different times — which is a less quotable answer and a more accurate one.

Does executive function predict life outcomes?

The most-cited evidence here comes from the Dunedin study, which followed 1,000 children in New Zealand from birth to the age of 32. Terrie Moffitt and colleagues reported in 2011 that childhood self-control predicted adult physical health, substance dependence, personal finances and criminal offending — and not as a threshold effect but “following a gradient of self-control”: each step up in childhood self-control went with better outcomes, across the whole range.

The detail that matters for anyone comparing this with IQ is that the authors explicitly report that the effects “could be disentangled from their intelligence and social class”, and from mistakes the participants made as adolescents. They then reported a second, separate cohort of 500 sibling pairs, where family background is shared: the sibling with lower self-control still had poorer outcomes.

Two cautions, because this study is routinely oversold. Self-control is one part of executive function, not the whole of it — it sits closest to the inhibition component. And a gradient across a population is not a prediction about any individual child. What the study supports is that this ability carries information about later life that an intelligence measure does not already contain, which is a real and useful claim on its own.

Can executive function be trained?

Partly, and less than the products built on the idea imply. The most thoroughly tested version is computerised working memory training, and a 2016 meta-analysis by Monica Melby-Lervåg, Thomas Redick and Charles Hulme pooled 87 publications with 145 experimental comparisons. Their findings split cleanly:

• Near transfer is real. Immediately after training there were reliable improvements on verbal and visuospatial working memory — the abilities closest to what was practised.
• Far transfer was not found. Against a treated control group there was “no convincing evidence of any reliable improvement” on nonverbal ability, verbal ability, word decoding, reading comprehension or arithmetic. Their publication-bias analysis concluded there was no evidential value in that set of studies at all.

One detail makes the result harder to argue with: how much a participant improved on the working memory task was not related to how much they improved on anything else. If the training worked by strengthening a general capacity, the people who gained most on the exercise should have gained most elsewhere. They did not. What broader evidence on raising cognitive performance looks like is covered in how to increase your IQ, and the mechanism people usually have in mind when they assume training must work is set out in neuroplasticity.

Executive function difficulties, ADHD and autism

Difficulty with executive function is common, it is not a measure of intelligence or of effort, and it is not in itself a diagnosis. It is described as a feature of several conditions rather than as a condition of its own — ADHD and autism are the two most often discussed — and it also changes with tiredness, stress, illness, grief and age. The same person's executive function is not a fixed quantity from one month to the next.

Because the three components are separable, difficulty is usually uneven rather than global. A person may find holding information in mind straightforward and switching approach hard, or the reverse. That is what the research would predict, and it is one reason a single overall label is a poor description of anyone.

Assessment of executive function in a clinical or educational setting is done by a qualified professional, using a combination of standardised tests, structured interviews and information about how the person manages in ordinary life — not by a single online test, and not from a checklist of traits. If you are wondering about this for yourself or someone else, a doctor, psychologist or your child's school is the right place to start.

Trust and scope notes

This page is educational. It summarises published research on executive function; it is not a psychological or medical assessment of anyone, and nothing on it is a diagnosis or advice about one.

Our IQ test measures reasoning performance. It does not measure executive function, it cannot detect, screen for or rule out any condition, and a score on it says nothing about a person's inhibition, flexibility or day-to-day self-regulation — which, as the 2006 study above shows, is a statement about intelligence tests in general and not a limitation peculiar to ours. What an IQ score does and does not cover is set out in how IQ tests work, and the different abilities people mean by “intelligence” in types of intelligence.

IQ Revealed is not affiliated with, endorsed by or connected to Pearson (publisher of the Wechsler scales), the publishers of the Wisconsin Card Sorting Test, the Center on the Developing Child at Harvard University, the Dunedin Multidisciplinary Health and Development Study, or any of the universities, journals or researchers named on this page. Test names are trademarks of their owners and are named only to describe them. Quoted phrases are taken from the published titles and abstracts of the works linked in the sources.