Growth Mindset: What the Research Shows About Belief, Learning and the Brain.

Does believing you can get smarter help you learn? That is the question behind growth mindset, the belief that ability can grow with effort and good strategy. The idea began with a paper that sorted people by how they see intelligence.^[1] Some, whom the paper called entity theorists, saw it as fixed: you have it or you don't. Others, the incremental theorists, believed it could grow. Nearly four decades on, the idea has been tested, criticised and reshaped, and the answer is less simple than the popular version.

Start with the size of the link. In an international survey of fifteen-year-olds, growth mindset was associated with higher reading scores on average, after allowing for students' and schools' socioeconomic background.^[3] Without that adjustment, the reading gap linked to mindset would look larger, because disadvantaged students hold fixed mindsets more often and background predicts scores on its own. A meta-analysis, which pools many studies into one estimate, combined more than a hundred studies of mindset and school achievement and found a small but reliable link.^[4] Mindset accounted for about one per cent of the differences in achievement between students.^[4]

One per cent sounds trivial until you look at who benefits. A national randomised controlled trial put this to the test. Ninth-graders in US public schools were assigned by chance to less than an hour of online growth mindset lessons or to a comparison group. Lower-achieving students gained about a tenth of a grade point.^[5] Their gain was larger in some school settings: lower- and middle-achieving schools whose peer norms backed the message.^[5] The lessons cost almost nothing to deliver.

The backlash deserves a hearing. A later meta-analysis of dozens of programme trials corrected for publication bias, the tendency of positive results to be published more often than null ones.^[6] After that correction, the average effect on achievement was tiny and not statistically reliable, meaning it could not be told apart from no effect.^[6] In the six best-designed trials it was essentially zero.^[6] Its authors argued that many earlier positive results may reflect flaws in method rather than real change.

Before that analysis, two researchers behind the field's main trials had argued in a review that effects depend on students and settings.^[7] The real question, they argued, is for whom growth mindset works, under what conditions, and how. A second review of theirs, covering decades of research, reached the same view. The effect is real but varies from group to group, and it is strongest for at-risk students in supportive settings.^[8]

The two sides still disagree. The later meta-analysis tested those differences between students and settings but found no reliable ones.^[6] The national trial found modest gains in particular groups of students and schools.^[5] Averaged over everyone, growth mindset barely registers.

Another meta-analysis looked at people's implicit theories of intelligence, their unspoken beliefs about whether intelligence can change. Overall, growth beliefs were linked to higher achievement, but the association varied by region and was not reliable everywhere.^[9]

While psychologists argued over effect sizes, how large a change really is, neuroscientists were recording brains. Two brain-wave studies found that people with a growth mindset show a larger brain response within half a second of a mistake.^[11][12] That finding does not settle the school debate. It raises a new question. Does the belief change how the brain handles errors, or do people whose brains react this way tend to hold the belief? The next section sets out what the brain studies can and cannot say.

The key moment comes within half a second of a mistake. In one small lab study, researchers recorded brain waves through scalp electrodes while adults did a task built to cause frequent errors.^[11] After each error, the brain produced a positive voltage wave, and this wave was larger in people who held a growth mindset.^[11] The wave, called the error positivity (Pe), marks conscious attention to the mistake. People with a larger Pe were more often right on the next try.^[11]

The Pe is an event-related potential: a voltage change tied in time to a mental event, here the error. Statistically, the size of the Pe helped explain the link between mindset and accuracy after errors.^[11] A later study in a larger group of school-age children also found a larger Pe with growth mindset, which makes a fluke of the small adult sample less likely.^[12] In the children, though, the Pe did not explain the link between mindset and accuracy.^[12]

That detail matters. The adult study involved only 25 people.^[11] In both studies, people reported their own mindset on a questionnaire; no one was assigned a mindset. So these studies cannot say which comes first. The belief may enlarge the Pe, or people whose brains react more strongly to errors may tend to adopt the belief.

Reviews of brain-imaging work discuss a few brain areas. One review named two of them. The dorsal anterior cingulate cortex (dACC) tracks conflict and errors. The dorsolateral prefrontal cortex (DLPFC) supports planning and control. The review describes links between growth mindset and resting connectivity involving these areas used for error monitoring and control.^[13] A later review found a small, varied body of brain-recording and imaging studies.^[14]

A brain-scan study used functional MRI, which tracks blood flow as a sign of brain activity, to measure people at rest. The striatum, deep in the brain, helps predict rewards. It compares outcomes with expectations and adjusts what we do next. At rest, growth mindset went with a distinct pattern of cortico-striatal circuitry: functional coordination between the striatum and the frontal networks that control thought and action.^[15] In the working model, the extra attention after an error feeds this system, which then pushes toward fixing the mistake rather than avoiding it.

A computer-modelling study offers one idea of how this could work. It proposes that dopamine, the brain's reward chemical, trains the front of the brain to become a learning system of its own, called meta-reinforcement learning.^[16] Put simply, the front of the brain learns how to learn. That study does not mention growth mindset, so the link to mindset is an inference.

A training study in children came at the question from the other side: it trained a skill and then measured the belief. In a controlled four-week training study, primary-school children who trained improved more in mindset scores than no-contact controls.^[17] Changes in brain activity and coordination across the dACC, the striatum and the hippocampus, a memory area, were linked to mindset gains.^[17] The brain-scan group was small.

The study does not establish that brain change caused belief change. Taken together, the brain-wave, resting-scan and training studies fit a two-way model, in which belief and brain change may feed each other.

The last step in the model is storage. One well-studied way learning is stored involves stronger connections between neurons.^[18][19] When neurons fire together again and again, the link between them grows stronger, a process called long-term potentiation (LTP).^[18] A growth protein, brain-derived neurotrophic factor (BDNF), supports dendritic spine growth (the small bumps where neurons connect), and it is essential for storing memories in the hippocampus.^[18][19] In the model, learning from a mistake would rely on the same process, called synaptic consolidation: connections strengthening in the circuits used to make the correction.

A tenth of a grade point sounds small, but context changes the picture. The lessons were cheap to deliver at scale. They helped lower-achieving students at the start of high school, a time when grades often begin to split apart.^[5] And in lower- and middle-achieving schools whose peer norms backed the message, those students gained more.^[5] Neuroplasticity, the brain's ability to change its structure and function with experience, also depends on context.^[19] That parallel is suggestive, but the school trial measured grades, not brains.

Seen through these studies, growth mindset looks less like a personality trait and more like a habit of attention. People with a growth mindset paid more attention to their errors, as the brain-wave studies showed. Whether they also store corrections more durably has not been measured; that step belongs to the model.