7 Thinking Skills: What the Research Shows, and How to Practise Each One

Start hereIn 25 experiments, people told to imagine another person's view were no more accurate about that person. In a final experiment, one thing did help: getting their view in conversation.

Seven thinking skills have research behind them. They are judging how sure to be, changing course, staying clear under pressure, understanding other people, curiosity, telling real patterns from chance, and updating your views. Below: what the studies found on each, how each relates to IQ where researchers measured it, and one small way to practise it.

An IQ score captures some kinds of thinking well and others only in part. A 1996 American Psychological Association task force wrote that standardized tests do not sample all forms of intelligence. Its examples included creativity, wisdom, practical sense and social sensitivity. In seven studies, many classic thinking biases were unrelated to cognitive ability, though some others were related.

Practice helps several of these skills, usually in narrow ways.

011. The seven at a glance

Each skill has research behind it, but practice mostly improves the exact thing you practise.

Skill	In plain words	What the studies we read found
Metacognition	Judging how well you know something	One lab study found training helped; a 2022 repeat did not
Cognitive flexibility	Switching approach when things change	In children, gains stayed with the skill that was trained
Emotion regulation	Thinking clearly while feeling strongly	Rethinking the situation helped; suppressing the feeling did not
Perspective-taking	Understanding what others think	Training helped young children; in one experiment, asking beat imagining
Curiosity	Wanting to find out	Memory improved for what people were curious about
Pattern recognition	Telling real patterns from chance	Reliable intuition needs a predictable setting and a chance to learn it
Intellectual humility	Updating your views on evidence	Short exercises raised people’s own ratings for a while; few have been repeated

022. The seven skills

Each section says what the skill is, what the research shows, how sure the evidence is, and one small way to practise it.

Metacognition: knowing how sure to be

Metacognition means judging your own thinking; whether lab training can sharpen it is still debated. It is awareness of your own thinking, such as knowing whether you really know an answer. For example, a word on the tip of your tongue is a common case: you know that you know it, but you can’t reach it yet.

In a 2019 study, people who got feedback on their confidence over eight sessions became better calibrated than a group given feedback only on their answers. The gain carried over to a memory task they had not trained on. A 2022 preregistered replication that removed two possible confounds found moderate evidence of no training effect. Its authors concluded that earlier claims about metacognitive training are premature.

For schools, the Education Endowment Foundation rates teaching metacognition and self-regulation as high impact for very low cost, based on extensive evidence. It estimates eight months of extra progress for pupils, on average.

In one small study of secondary-school students, the quality of their metacognitive skills predicted maths results on top of intelligence among the older group. In that study, it added something the intelligence score did not show.

One brain-imaging study here shows a link; three others tested cause, with mixed results. A 2010 imaging study found that people who judged their own perceptual decisions more accurately had more grey matter in the anterior prefrontal cortex. In a 2014 study, seven patients with damage there judged their perceptual decisions less accurately, but judged their memory normally.

A 2010 study that dampened another frontal region with magnetic stimulation impaired these judgements. A 2017 attempt to repeat it found no impairment in two experiments. With one result and one failed repeat, that question is still open.

How sure: Unsettled a lab training gain did not appear in a preregistered repeat; a school review rates teaching it highly; brain studies are few.

Try it (our suggestion): before you check an answer, note how sure you are: a guess, fairly sure, or certain. After a week, see how often your “certain” answers were right.

Cognitive flexibility: changing course

Cognitive flexibility is switching approach when things change; training it mostly improves the trained task. It is one of the executive functions, the mental skills you use to plan, focus and hold back impulses. For example, when your usual route is closed, flexibility is finding another way rather than waiting at the barrier.

A 2013 review lists the core executive functions as inhibition, working memory and cognitive flexibility. It describes flexibility as including seeing anything from different perspectives and quickly adapting to changed circumstances.

A 2019 meta-analysis found that training one of children’s executive functions improved that skill. It found no convincing evidence that the gains spread to the skills that were not trained. Researchers call that kind of spread far transfer.

In 2017 Giovanni Sala and Fernand Gobet reviewed studies of chess, music and working-memory training. The better the experimental design, the smaller the effect. They “conclude that far transfer of learning rarely occurs”. The weight of the evidence we read leans towards far transfer being rare, with action games and music training the main exceptions claimed. Our Range summary covers that debate.

In a 2006 study of young adults, updating working memory was strongly related to intelligence scores. Shifting between mental tasks was not, once the other skills were accounted for. Its authors suggested that intelligence measures do not equally assess every control ability.

How sure: Moderate children’s gains stayed with the trained skill; far transfer is still argued, and the evidence we read leans against it.

Try it (our suggestion): once a week, solve a familiar problem a second way. Expect it to help with that kind of problem; in the studies, gains stayed close to what was practised.

Emotion regulation: thinking clearly under pressure

How you handle a strong feeling matters; rethinking the situation tended to work better than pushing the feeling down. Emotion regulation is how you manage your own feelings, on purpose or without noticing. For example, reading a curt email as “they’re rushed” rather than “they’re angry with me” changes how you feel about it.

A 2012 meta-analysis of 306 experimental comparisons found that changing how people thought about a situation had a small-to-medium effect on their emotions. Suppressing the outward expression of an emotion helped, but trying to suppress the feeling itself did not. Among ways of rethinking, taking a different perspective did better than reappraising your own reaction.

Stress changes how well you think in the moment. A 2016 meta-analysis found that acute stress impaired working memory and cognitive flexibility, with more nuanced effects on inhibition. Its authors described a stressed state with more reactive processing of salient things but greater control over actions. A 2009 review stated that even quite mild acute uncontrollable stress can cause a rapid and dramatic loss of prefrontal cognitive abilities.

These are lab findings about single moments. If strong feelings often overwhelm you or get in the way of daily life, talk to a doctor or a mental health professional. Our overthinking guide describes when worry is worth that conversation.

How sure: Moderate a large meta-analysis supports rethinking over suppressing the feeling, but these are lab tests of single moments.

Try it (our suggestion): when you notice a strong feeling before a decision, write down one other way to read what happened. Then decide.

Perspective-taking: understanding other minds

In one experiment, asking someone improved accuracy about their thoughts; in 25, imagining their view did not. The ability behind it is called theory of mind: understanding that other people have beliefs, wants and feelings different from your own.

In 25 experiments published in 2018, people told to imagine another person’s perspective did not judge that person’s thoughts or feelings more accurately. If anything, accuracy fell overall, while confidence sometimes rose. In a final experiment, getting the other person’s perspective through conversation did increase accuracy.

For young children, a 2016 meta-analysis of 45 studies found that theory-of-mind training worked better than control activities. A 2015 analysis of a popular test of reading feelings from photographs of eyes found its short form strongly related to vocabulary. Our reading: measuring social understanding apart from word knowledge is hard.

Brain studies point to a network used for this, and researchers still debate how specialised it is. A 2003 imaging study reported a region where the temporal and parietal lobes meet that was involved specifically in reasoning about another person’s mind. A 2014 meta-analysis found overlapping activity across theory-of-mind tasks in two areas, supporting the idea of a core network. A 2008 study found that the right-side region also responded to a task with no social content that required redirecting attention. A 2009 study with sharper scans found neighbouring but distinct regions there, one for theory of mind and one for shifting attention.

For example, you may feel sure you know why a colleague is upset. Asking “what’s bothering you most?” tests that guess.

How sure: Moderate imagining was tested in 25 experiments and asking in one final experiment, all by one team; the brain findings are still debated.

Try it (our suggestion): swap one guess a day for a question, such as “what did you mean by that?”

Curiosity: wanting to find out

Curiosity helps you remember what you were curious about; its effect on unrelated things is mixed. For example, the answer to a trivia question you badly wanted to know tends to stick.

In a 2009 study, people recalled surprising trivia answers better one to two weeks later when they had been more curious. In the same work’s brain scans, curiosity about trivia questions correlated with activity in regions previously suggested to be involved in anticipated reward.

A 2014 study found better memory for answers people were curious about, and for unrelated faces shown while they waited. It found more activity in the midbrain and nucleus accumbens during high curiosity. A 2021 study found the boost for unrelated faces only when they appeared shortly after the question. In three 2023 experiments on guessing before seeing an answer, which some theories link to curiosity, people remembered the answers better but not the unrelated material. In a 2024 study, people remembered unrelated school facts worse when they came after a trivia question that had made them very curious.

A 2011 analysis found intelligence the single strongest predictor of academic performance, but curiosity and effort together rivalled it. So curiosity added something an intelligence score did not. The studies we read raised or measured curiosity in the moment; they did not train it as a lasting trait.

How sure: Moderate the boost for answers is consistent; for unrelated things it is mixed, and one study found worse recall of school facts.

Try it (our suggestion): before you read about something, write one question you want answered, and guess the answer first.

Pattern recognition: real patterns or chance

Spotting patterns is part of some IQ tests and of expertise, but people also see patterns that aren’t there. For example, three wins in a row can feel like a hot streak; the useful question is how often that would happen by luck alone.

A 1990 analysis of Raven’s Progressive Matrices, a widely used nonverbal intelligence test, looked at what set higher scorers apart. It was mainly the ability to work out abstract relations and to keep track of many problem-solving goals in working memory. So this skill is, in part, what some IQ tests measure.

A 1996 study supported the usual estimate that chess masters store some 50,000 chunks, familiar patterns of pieces, in memory. A 2009 model showed that natural selection can favour links that are often wrong, as long as the occasional correct response carries a large benefit. Our reading: that helps explain superstitions.

In 2009 Daniel Kahneman and Gary Klein concluded that judging an intuition means asking two things. Is the environment predictable, and has the person had the chance to learn its regularities? They add that “subjective experience is not a reliable indicator of judgment accuracy”.

One theory treats the brain as built around prediction. In 2013 the philosopher Andy Clark called the hierarchical prediction machine approach the best clue yet to a unified science of mind and action. The same article examined its pitfalls and challenges, including evidential ones. Critics argued in 2020 that its models are seldom empirically validated and that calling it a unifying theory is premature.

How sure: Moderate the test and chess findings come from detailed lab studies; the evolution model and the prediction approach are theories still being tested.

Try it (our suggestion): write down a prediction before you find out, such as how long a task will take, then check. Our correlation vs causation guide helps when two things seem linked.

Intellectual humility: updating on evidence

Intellectual humility is recognising that your beliefs might be wrong; brief exercises raised it, but few have been repeated. Researchers at Duke University defined it as “the degree to which people recognize that their beliefs might be wrong”. A 2022 review found that the common element across definitions was a metacognitive ability to recognise the limits of one’s beliefs and knowledge. That makes it a close cousin of the first skill on this list.

A 2019 study found that both cognitive flexibility and intelligence predicted intellectual humility, but neither was necessary.

The 2022 review describes short exercises that raised it for a while, such as reflecting on a situation from the view of a distant observer. In another, people who first listed two things they did not know reported less confidence in their answer. Most of these experiments used small to medium samples and have not been replicated; the distant-observer effect is one exception.

Feelings may play a part in changing your mind. In a 2016 imaging study, 40 liberals read arguments against their political and non-political views. Those who changed their minds more showed less activity in the insula and amygdala as they weighed the counterevidence. The authors said the results highlight the role of emotion in resisting belief change.

For example, before a family debate about a news story, you might notice you have only read one report of it. Our Think Again summary looks at rethinking in more depth.

How sure: Limited the exercises raised it briefly in mostly small studies, few have been repeated, and the brain finding is one study of 40 people.

Try it (our suggestion): before an argument or a big decision, list two things you don’t know about it.

033. Putting it together

Pick one skill, practise it for a week, and check whether anything changed. The seven tries from this page:

• Metacognition: note how sure you are before you check an answer.
• Flexibility: solve one familiar problem a second way each week.
• Emotion regulation: write one other way to read what happened before you decide.
• Perspective-taking: swap one guess about someone for a question.
• Curiosity: write a question and a guess before you read.
• Patterns: write down predictions, then check them.
• Intellectual humility: list two things you don’t know before you argue.

Our How to Learn Faster guide covers the study habits behind several of these. Our cognitive biases page looks at famous thinking-error findings, including the Dunning-Kruger effect.

04Quick answers

05Use this

06Quick quiz

How we checked this · Last checked October 2026

• We read the abstract of every study through PubMed, OpenAlex, Europe PMC or the publisher's page, and the full text of the 2022 intellectual-humility review.
• We read the 1996 APA task force's summary of findings from a copy of that summary, and cite it by its DOI.
• Brain-imaging results are described as links, not causes. Where a brain-injury or stimulation study tested cause, we say so, and we show the 2017 study that did not repeat a stimulation result.
• The 2019 metacognition training study is shown beside the 2022 preregistered replication that did not find the effect.
• Our wording on far transfer and on judging intuitions matches our Range summary.
• A 2024 study that found worse memory for unrelated school facts is shown beside the earlier curiosity studies.

How we check every article →

07Sources