Two Kinds of Mental Flexibility Use Different Brain Circuits

What the research describes

The Wisconsin Card Sorting Test, in use since 1948, hands a person cards to sort by colour, by shape, or by number of objects. Feedback from the examiner is the only guide to which rule is correct, and once the sorting has become reliable the examiner changes the rule without saying so. Children under four and adults with frontal lobe damage tend to persist with the rule that has stopped working.

A reversal learning task asks something narrower. Two choices are each paired with an outcome, the pairing is learned across a series of trials, and then the two outcomes are swapped. No rule governing the situation has changed, and the single association just formed now runs backwards.

Lucina Uddin, a neuroscientist at the University of Miami, drew the human and animal literatures together in a 2021 review for the journal Nature Reviews Neuroscience. Both tasks appear in it, both are routinely described as measures of flexibility, and one of the review’s through-lines is that they call on separable systems. The article is a narrative review rather than a meta-analysis, with no stated search protocol and no pooled statistics, so it summarises a field rather than testing it.

The dissociation is clearest in lesion work with marmoset monkeys. Animals with damage to the lateral prefrontal cortex were impaired when the relevant feature of a stimulus moved from one category to another, from colour to shape, while their reversal learning survived intact. Monkeys with damage to the orbitofrontal cortex showed the opposite profile. Inactivating the orbitofrontal cortex in rats impairs reversal learning specifically, through persistence with the choice that used to pay.

Human imaging finds comparable anatomy, and for people a reversal is far easier to manage than a shift between categories. A third sense of the word lives outside the laboratory: the adjustment a child makes when a routine changes, a plan collapses, or a lesson takes a turn nobody announced. Questionnaires completed by parents and teachers index that behaviour, and they correlate only weakly with performance on a card sort. Uddin’s account of the discrepancy is that the two formats interrogate different situations, one a structured task with an examiner present, the other an ordinary unstructured day.

What this means for your child

The review gives each clinical condition its own manifestation of inflexibility. In autism it appears as restricted and repetitive behaviour, including insistence on sameness, and the severity of those behaviours is associated with measures of cognitive inflexibility. In obsessive-compulsive disorder it appears as recurrent and persistent thoughts, alongside reduced orbitofrontal and frontostriatal activation during flexibility tasks. Depression and generalised anxiety are both entered under repetitive negative thinking, schizophrenia and the later-life dementias under excessive rigidity.

ADHD occupies an odd position on the same table, listed under inattention rather than rigidity. Uddin observes that the diagnostic criteria can be read as distractibility, or as too much flexibility, and then declines the tidy contrast, because autism and ADHD co-occur often enough that impaired shifting and inattention frequently arrive in the same child. Where the two groups have been compared directly, children with autism performed worse on the card sort than children with ADHD.

Executive function is compromised in nearly every DSM-5 category, which leaves open how much diagnostic information a flexibility deficit carries on its own. Only a handful of imaging studies have examined autism and ADHD together. One reported shared and distinct patterns of connectivity, another reported no group differences at all.

A low score on a card sort records what happened when an examiner changed a rule without announcing it. The review states plainly that well-documented inflexible everyday behaviour in autism is not necessarily related to cognitive inflexibility as assessed experimentally, so a child who cannot leave the house when the swimming lesson is cancelled and a child who persists with a rule that has stopped working may both be called inflexible, and no available measure separates them.

Rigidity is also, on this evidence, sometimes the appropriate setting. Cognitive stability serves focused attention and the ability to ignore a distraction, and Uddin raises the possibility that some reductions in flexibility are reasonable adaptations to an environment the person experiences as stable.

What the field lacks is a standardised, transdiagnostic assessment, normed by age, and until one exists no score can say which of these difficulties a particular child has. Uddin names the development of such an instrument as unfinished work, along with the unresolved problem of connecting laboratory scores to behaviour outside the testing room. The word came into your child’s report carrying more confidence than the science behind it can currently support.