Can You Improve Your IQ? What the Research Actually Says

Want to know your current score before trying to improve it? You can take our 50-question assessment and get your IQ with a percentile breakdown across four cognitive domains — a useful baseline for tracking any changes over time.

Is IQ fixed, or can it change?

The idea that IQ is sealed at birth is one of the most stubborn myths in popular culture, and it's wrong. The reality is more interesting. IQ is one of the most stable psychological traits ever measured — the classic Scottish Mental Survey found a correlation of 0.66 between the same test taken at age 11 and age 77. But “stable” is not the same as “fixed.”

Twin and adoption studies consistently put the heritability of adult IQ at 50-80%. That sounds like a ceiling, but it doesn't mean what most people think. Heritability measures variance within a population— not the proportion of any individual's IQ that's genetic. An individual whose environment changes dramatically can see real, measurable shifts in IQ over time. The question isn't whether IQ can change; the question is which interventions produce reliable, durable, transferable changes.

The honest picture from the research: modest improvements are achievable through well-supported interventions. Large, permanent leaps are not. Understanding the difference is what separates effective effort from expensive disappointment.

What actually works, ranked by evidence

Here's the research-backed hierarchy, from best-supported to most oversold. Note that the biggest effects come from interventions that take years, not weeks.

1. Formal education (strongest evidence)

A 2018 meta-analysis by Stuart Ritchie and Elliot Tucker-Drob at the University of Edinburgh synthesised 28 studies covering over 600,000 participants. Their finding: each additional year of formal education raises measured IQ by 1 to 5 points, with an average around 3.4 points per year.

A large-scale Danish study (Hegelund et al., 2020) replicated this, finding roughly 4.3 IQ points per year of education in young adulthood and 1.3 points in mid-life. Critically, the effects were strongest for people who started lower on the distribution — meaning education has a partial equalising effect on measured cognitive ability.

This is the clearest finding in the literature. If you can add years of substantive formal study to your life, you will, on average, raise your measured IQ. The mechanisms are debated (crystallised knowledge? reasoning practice? test-taking familiarity?), but the effect is real and replicated.

2. Physical exercise (strong evidence)

Regular aerobic exercise produces measurable cognitive benefits. MRI studies show that sustained exercise promotes neurogenesis in the hippocampus, increases brain-derived neurotrophic factor (BDNF), and improves cerebral blood flow. The effect on measured IQ is modest — usually 2-4 points in intervention studies — but it's robust across populations.

Combined programmes (aerobic + strength + coordination) tend to produce larger effects than any single modality. The benefits appear at all ages but are especially pronounced in children and older adults. If you're looking for one lifestyle intervention with real cognitive upside, this is it.

3. Sleep quality (strong evidence)

Sleep isn't just a “nice to have” for cognition; it's central to it. Research on sleep deprivation shows that a single night of poor sleep can depress working memory performance by up to 10 IQ points. Over the long term, chronic sleep debt is associated with measurably lower cognitive function.

The flip side: consistent good sleep (7-9 hours, regular schedule) produces a meaningful but usually invisible benefit — you don't gain IQ points from sleeping well, you just stop losing them. If your tested score is being held down by poor sleep hygiene, this is one of the easiest wins available.

4. Learning genuinely new skills (moderate evidence)

Active engagement with demanding new material — learning a language, taking up a musical instrument, learning to code, serious study of a novel domain — correlates with cognitive maintenance and, in some studies, modest IQ gains. The effect appears strongest when the learning is difficult and sustained. Easy tasks don't produce the effect.

The mechanism is probably what researchers call “cognitive reserve” — a richer network of neural connections built through challenging use. It won't transform your IQ, but combined with the other evidence-backed factors, it contributes.

5. Diet and nutrition (moderate evidence)

A diet providing adequate omega-3 fatty acids, B vitamins, iron, and iodine supports cognitive development and maintenance. Correcting significant deficiencies can produce meaningful IQ increases — particularly in children or during pregnancy. In well-fed adults with no deficiencies, marginal dietary tweaks produce minimal effect. The big wins are at the floor, not the ceiling.

6. Extended intensive cognitive training (weak-moderate evidence)

Here's where the picture gets genuinely contested. A 2020 reanalysis of Kvashchev's long-term study found that 3 years of intensive creative problem-solving training produced IQ gains of 8-10 points. Studies on dual n-back training have produced gains of 3-4 points in fluid intelligence, though these results are fiercely debated and don't replicate consistently.

Where there is agreement: extended (months to years) and intensive cognitive training can produce modest, real gains. Short interventions (weeks) rarely do, and gains often fade when training stops. If you're willing to commit to a serious cognitive programme for a year or more, there's plausible evidence of real benefit. If you're looking for a quick boost, this isn't it.

What the research does NOT support

The cognitive-enhancement industry is enormous, and much of it is built on claims the evidence doesn't support. Three categories are particularly worth calling out.

Commercial brain-training apps

Lumosity, Elevate, Peak, CogniFit, and similar apps have been studied extensively. The consistent finding is that they improve performance on the specific games but rarely transfer to general intelligence. A 2016 consensus statement signed by 70+ cognitive scientists concluded that the evidence does not support claims of lasting, generalisable cognitive enhancement from these apps.

Lumosity itself paid $2 million to settle FTC charges of deceptive advertising in 2016. Meta-analyses show that when studies include proper placebo controls — a key methodological safeguard — the effects of brain-training apps on IQ shrink to near zero.

Playing these games is fine; they're mildly entertaining and the worst they'll do is waste your time. But expecting IQ gains is not supported by the research.

Practice tests and coaching

Taking an IQ test repeatedly produces a practice effect — your second score is typically 3-8 points higher than your first, and subsequent attempts smaller bumps after that. This isn't intelligence gain; it's familiarity with the test format. Clinical psychologists explicitly wait at least a year between administrations of the same test because of this.

So you can “improve” your score on a specific test by practicing — but only on that test, and the effect plateaus quickly. If you need to impress on a specific IQ test (e.g., qualifying for a gifted programme), some practice with similar formats helps. But it doesn't make you measurably smarter at anything else.

Supplements and “nootropics”

The evidence for most supplements marketed for cognitive enhancement is weak to non-existent in healthy adults. Caffeine has small, real effects on alertness and reaction time. Fish oil may help if you're deficient in omega-3s. Creatine shows promising but limited effects in specific conditions. Everything beyond that — nootropic stacks, “smart drugs,” cognitive enhancer blends — has essentially no peer-reviewed support for lasting IQ gains.

This is an industry built largely on placebo, anecdote, and unpublished studies. The money is better spent elsewhere.

What's a realistic ceiling?

Adding up the evidence-backed effects, a reasonable estimate of what's genuinely achievable through sustained effort:

The key word is sustained. Every intervention that works tends to fade without continued effort. Ritchie and Tucker-Drob found that education gains persist largely because education itself is sustained — it changes how you spend your time, what you're exposed to, and what cognitive demands you face every day.

Beyond about 15 points, you're probably running into your genetic ceiling — the upper bound of what your neurobiology will support. This isn't a reason for despair; it's useful information. A 118 who works sustainedly can probably measure as a 125 within a few years. A 118 who expects to measure 135 through effort alone is going to be frustrated.

Measurement matters more than most people realise

One under-appreciated point: a lot of what feels like IQ change is actually measurement change. Every IQ score carries a confidence interval of roughly ±5 points. So a person whose true IQ is 117 might measure 112 one day and 121 another — with no change in underlying ability at all.

If your first test score was on a short online quiz (unreliable) and your second is on a substantial assessment (more reliable), the difference is mostly measurement, not change. Before chasing “improvement,” it's worth establishing a reliable baseline. See our article on IQ score accuracy for a deeper look at why measurement error matters so much for interpreting any “improvement” you think you're seeing.

This also means: if you take the same test twice and your second score is 5 points higher, that might just be measurement noise and practice effect. To detect a genuine gain from an intervention, you want the improvement to exceed the measurement error — typically meaning 8+ points across different tests, with reasonable time gaps between them.

A practical plan if you want to improve your IQ

Based on the research, here's an honest, evidence-backed programme if you're serious about raising your measured IQ. It's not glamorous, but it's what actually works.

1. Establish a reliable baseline. Take a substantial IQ test once, under good conditions (rested, fed, low stress). This is your starting point — not a short online quiz that could be off by 15 points.
2. Fix the basics first. Sleep 7-9 hours consistently, exercise 150+ minutes weekly (mix of aerobic and strength), eat adequately. These aren't IQ-booster hacks; they're the foundation everything else builds on.
3. Engage in demanding new learning. A language, a musical instrument, programming, a genuinely difficult subject. Easy tasks don't produce the effect. You want to be regularly uncomfortable — that's what builds cognitive reserve.
4. If possible, pursue formal education. This is the single-largest evidence-backed intervention. Completing a degree, an advanced qualification, or substantive professional training will produce measurable gains over years.
5. Read widely and think actively. Not just passive consumption — deliberately engage with ideas, write about them, debate them. Crystallised intelligence (accumulated knowledge and reasoning patterns) is the component of IQ that continues to grow through adulthood.
6. Retest after 6-12 months minimum. Earlier testing is mostly measurement noise + practice effect. Wait long enough that genuine changes can accumulate and be detected above the confidence interval.

What's notably absent: no brain-training apps, no supplement stacks, no weekend intensives. Those aren't forbidden — they're just not what the research actually supports.

Should you bother trying?

An honest question worth asking. IQ correlates with life outcomes — job performance, income, educational attainment — but the correlations are real rather than dominant. Most of the variance in those outcomes comes from everything else: conscientiousness, opportunity, network, luck, domain expertise, and sustained effort. A 5-point IQ gain is worth something, but probably less than an equivalent investment in a specific skill or career development.

Where IQ improvement is clearly worth pursuing: if you're at a life stage where education is natural (still in school, considering further study, changing careers), the cognitive gains come as a bonus to the primary benefits of the education itself. If you're looking to maintain cognition in older age, the lifestyle interventions (exercise, sleep, learning) are worth doing anyway for broader health reasons.

Where it's probably not worth pursuing: chasing a specific IQ threshold (e.g., trying to qualify for Mensa by raising a score from 125 to 132) through effortful intervention. The measurement error on any single test is usually larger than the gain you'll produce through short-term work. You're better off taking multiple high-quality tests — if your true ability is near the threshold, one of them will probably reflect it.

The bottom line

Can you improve your IQ? Yes, within limits. The interventions that work are boring but effective: formal education, physical exercise, good sleep, genuinely challenging cognitive engagement. Sustained over years, these can produce meaningful gains — plausibly 5-15 points depending on your starting point and commitment.

The interventions most heavily marketed — brain-training apps, supplements, quick-boost programmes — have little to no evidence of producing lasting, transferable cognitive enhancement. You can skip them without missing anything.

And before focusing on improvement, make sure you have a reliable measurement of where you are. The difference between a 112 on a short online quiz and a 117 on a substantial assessment is probably just measurement quality, not genuine change. Get a good baseline first, then work on the factors that genuinely matter.

Frequently asked questions