The short answer
IQ is not fixed. Under certain conditions — particularly in childhood and adolescence — the environmental contributions to IQ are large, and interventions can produce meaningful gains. The Flynn Effect (average IQ rising ~3 points per decade globally throughout the 20th century) is proof that IQ responds to environmental change at the population level.
In adulthood, malleability is more limited. Most commercial brain training products produce no lasting improvement in general cognitive ability — they improve performance on the trained tasks, which does not transfer to untrained tests or real-world performance. The things that actually move the needle are unsexy: education, sleep, exercise, and nutrition. None of them are sold in app stores.
There is also a critical distinction between improving performance on IQ tests (achievable through familiarity with test formats, gaining ~5–10 points) and improving the underlying cognitive abilities those tests measure (much harder, where most interventions fail at far-transfer).
Evidence table: every major intervention rated
Below is every major intervention claimed to improve IQ or cognitive performance, with an honest assessment of the evidence quality, effect size, and whether the gains transfer to general intelligence or just trained tasks.
| Intervention | Evidence quality | Effect size | Est. IQ gain | Transfers to general IQ? | Key caveat |
|---|---|---|---|---|---|
| Formal education | Strong | 1–5 pts / year | ✅ Yes — broad transfer | Effect strongest in childhood; diminishes in adulthood | |
| Correcting iodine deficiency | Strong | 10–15 pts | ✅ Yes — real ability gain | Only applicable where deficiency exists; no effect in sufficient populations | |
| Correcting iron deficiency (children) | Strong | 4–5 pts | ✅ Yes — real ability gain | Effect only in anaemic children; no effect in iron-sufficient individuals | |
| Aerobic exercise (sustained) | Moderate | 2–4 pts | ⚠️ Partial — exec. function & WM | Strongest in children and older adults; smaller effect in healthy young adults | |
| Sleep optimisation | Moderate | 2–5 pts | ⚠️ Removing impairment, not adding ability | Restores performance lost to sleep debt; not a true IQ increase above rested baseline | |
| Reading extensively | Moderate | 1–3 pts (crystallised) | ⚠️ Crystallised only — not fluid IQ | Builds vocabulary and Gc; limited effect on pattern-reasoning Gf component | |
| Stress reduction / mindfulness | Weak | 1–2 pts | ⚠️ Indirect — removes cognitive load | Chronic stress impairs working memory and prefrontal function; reducing it restores capacity | |
| Musical instrument training | Weak | 0–2 pts | ❌ Limited transfer | Earlier studies overestimated; recent RCTs show near-zero transfer to untrained tasks | |
| Chess training | Weak | 0–2 pts | ❌ Minimal transfer | Sala & Gobet (2017) meta-analysis found small effects that did not survive publication bias correction | |
| N-back training | Weak | 0–2 pts (near-transfer only) | ❌ Near-transfer only | Improves n-back performance; transfer to fluid IQ is contested and likely small | |
| Commercial brain training apps (Lumosity, BrainHQ etc.) | None | ~0 | ❌ No transfer demonstrated | 2014 consensus statement (75 scientists); Lumosity FTC fine 2016; improves trained tasks only | |
| Omega-3 / fish oil supplements | None | ~0 in sufficient populations | ❌ No effect in RCTs | Inconsistent results; no reliable effect in well-nourished adults or children | |
| IQ test practice / familiarity | Strong | 5–10 pts on score | ❌ Score only — not real ability | Improves test score without improving underlying ability; gains do not transfer to novel tasks |
What actually works — in detail
Education: the strongest lever
The most powerful IQ improvement tool available is sustained formal education. A 2018 meta-analysis by Ritchie and Tucker-Drob analysed 42 studies using methods that control for selection effects — particularly studies exploiting natural experiments like school entry cutoff dates (where children born days apart get a full year difference in schooling due to administrative cutoffs) and changes in compulsory schooling laws. Their finding: each additional year of formal schooling is associated with 1–5 IQ points, with a causal estimate of approximately 1–3 points per year.
Education works because IQ tests directly measure abilities that formal schooling develops: abstract reasoning, pattern recognition, working memory for complex problems, and verbal comprehension. The more you engage with structured problem-solving over time, the more your scores on these dimensions improve. This is not "teaching to the test" — it is genuine cognitive development.
The implication for adults is clear: continued learning — through further education, intellectually demanding careers, or systematic self-study — sustains and gradually builds cognitive ability in ways that passive entertainment does not.
Exercise: the best non-educational intervention
Aerobic exercise is the most robustly evidenced non-educational intervention for cognitive performance in healthy adults. Meta-analyses consistently find that regular aerobic exercise improves executive function (planning, cognitive flexibility, inhibitory control), working memory capacity, and processing speed, with small-to-moderate effect sizes (typically d = 0.3–0.5 for executive function).
The biological mechanisms are well-understood. Aerobic exercise increases production of BDNF (brain-derived neurotrophic factor), a protein that supports neuronal survival and promotes the growth of new neurons in the hippocampus — a structure critical for memory and learning. Exercise also improves cerebrovascular health, increasing blood flow to the prefrontal cortex during demanding cognitive tasks, and reduces chronic neuroinflammation that impairs cognitive function.
- Duration: 20–45 minutes per session
- Frequency: 3–5 sessions per week (150–200 min/week total)
- Intensity: Moderate — you should be able to speak but not sing (approximately 60–75% maximum heart rate)
- Timeline: Measurable cognitive effects typically emerge after 4–8 weeks of consistent training
- Best evidence for: Executive function, working memory, processing speed. Effects on fluid IQ directly are present but smaller.
Sleep: removing the largest common impairment
Chronic sleep restriction is probably the most widespread cause of below-optimal cognitive performance in developed countries. Research by Van Dongen and colleagues (2003) found that sleeping 6 hours per night for two weeks produced cognitive impairments equivalent to 24 hours of total sleep deprivation — and critically, participants were unaware of how impaired they were. They adapted to feeling only slightly sleepy while their performance continued to decline.
This is not technically "improving IQ" — it is removing a chronic impairment and allowing your actual cognitive capacity to express itself. But for the large proportion of adults chronically sleeping under 7 hours, optimising sleep is the single fastest route to a measurable improvement in cognitive test performance.
- Consistency: Same bedtime and wake time including weekends. Irregular sleep timing disrupts circadian rhythm and reduces sleep quality even at adequate duration.
- Temperature: Sleep environment 16–20°C (60–68°F). Core body temperature must drop ~1°C to initiate sleep; a cool room accelerates this.
- Light: Blackout curtains or sleep mask. Light exposure during sleep fragments sleep architecture and reduces slow-wave and REM stages.
- Screens: Avoid bright screen light for 60–90 minutes before sleep. Blue light suppresses melatonin secretion and delays sleep onset.
- Caffeine: Half-life is approximately 5–6 hours. Afternoon caffeine (after 2pm for most people) measurably reduces slow-wave sleep.
Nutrition: deficiency correction vs supplementation
The nutritional evidence for IQ improvement is straightforward but frequently misrepresented. There are two completely different situations: correcting a deficiency that is actively suppressing cognitive function, and supplementing beyond adequacy in the hope of further gains.
Correcting deficiencies produces large, real gains. Iodine deficiency is the world's leading preventable cause of cognitive impairment. Studies of iodine supplementation in deficient populations find gains of 10–15 IQ points (Bleichrodt & Born, 1994). Iron deficiency anaemia in early childhood impairs cognitive development; iron supplementation in anaemic children shows gains of 4–5 IQ points. These are genuine improvements in underlying cognitive ability, not just test score artefacts.
Supplementing beyond adequacy does not work. In well-nourished populations, omega-3 supplements, multivitamins, and most other cognitive "nootropics" show no reliable IQ gains in randomised controlled trials. The brain is not waiting for additional micronutrients — it already has what it needs. The supplement industry profits from confusing these two very different situations.
- Iodine: use iodised salt or eat iodine-rich foods (seafood, dairy). Especially critical during pregnancy.
- Iron: regular red meat, legumes, or supplementation if anaemia is diagnosed (requires blood test).
- Vitamin D: most people in northern latitudes are deficient. Get levels tested; supplement if below 50 nmol/L.
- B12: essential for nerve function; vegans should supplement. Deficiency causes cognitive impairment.
- Alcohol: even moderate regular consumption measurably reduces cognitive performance and accelerates brain volume loss with age.
- Chronic stress: sustained high cortisol shrinks hippocampal volume and impairs working memory. Managing stress is a legitimate cognitive intervention.
What doesn't work: the myth of brain training apps
The commercial brain training industry generates billions of dollars annually by exploiting the gap between what people hope is possible and what the research actually shows. Understanding why it doesn't work requires understanding the near-transfer vs far-transfer distinction.
Near transfer means your performance improves on tasks very similar to what you practised. This always happens. If you practise n-back tasks for weeks, you will get better at n-back tasks. If you use Lumosity's memory games daily, your scores on those games will rise. This is unremarkable — it is just practice effects.
Far transfer means that improvement in one cognitive domain transfers to unrelated tasks that measure the same underlying ability in a different format. This is what matters for general IQ — and this is where brain training consistently fails. A 2014 open letter signed by 75 cognitive neuroscientists concluded there was "no compelling scientific evidence" that brain training improves general cognitive ability.
Lumosity was fined $2 million by the US Federal Trade Commission in 2016 for making deceptive claims that their games would reduce cognitive impairment from Alzheimer's disease and other conditions. The consent order required them to stop claiming their games improve real-world performance without competent and reliable scientific evidence.
The situation with chess training and musical instrument training is similar. Earlier studies suggested these activities improved general cognitive ability; better-designed studies using active control groups have repeatedly failed to replicate these findings. Improvement stays local to the practised domain.
Test practice: real score gains, not real IQ gains
Practising IQ-style questions does improve IQ test scores, typically by 5–10 points on average over multiple practice sessions. This is well-established and not in dispute. But it is important to understand what this represents.
Test practice reduces anxiety, builds familiarity with question formats, and develops strategies for specific item types (e.g. recognising that matrix reasoning patterns follow predictable rule structures). These gains do not transfer to novel tests measuring the same abilities in different formats — the underlying g factor does not increase. What changes is your ability to demonstrate your true ability on that specific test format, not your actual ability.
For the purposes of achieving a better score on a specific test (an employment assessment, an admission process), test practice is effective and legitimate. For understanding your true cognitive ability, test practice inflates the score relative to what a naive test would show.
Realistic expectations for adults
For adults in developed countries with adequate nutrition, sleep, and education, the realistic scope for general IQ improvement is modest. Based on the evidence above, a reasonable upper estimate for cumulative gains through sustained effort across lifestyle factors:
Optimising sleep (removing chronic deprivation) could recover 2–5 points of cognitive performance. Sustained aerobic exercise over months to years might contribute another 2–4 points on executive function and processing speed measures. Continued formal education or intellectually demanding learning could add 1–3 points per year of engagement. Total realistic gain for an already adequately nourished and educated adult: approximately 5–12 points through sustained lifestyle optimisation over 1–2 years.
This is meaningful — the difference between functioning at 90% and 100% of your cognitive capacity matters for real-world performance. But it is a far cry from the dramatic gains marketed by supplements and apps. The most important framing: you cannot upgrade your cognitive hardware, but you can ensure the hardware you have is running in optimal conditions. That gap between sub-optimal and optimal matters more than most people realise.
For a deeper understanding of what fluid vs crystallised intelligence are and why they respond differently to intervention, see Fluid vs Crystallized Intelligence. For how IQ changes across the lifespan, see Average IQ by Age. And to get an honest baseline measure of your current cognitive performance, the AurorIQ test is free and uses IRT scoring — no email required.
Frequently asked questions
Can adults increase their IQ?
Yes, but modestly. Adults with adequate nutrition, sleep, and education have less room for large IQ gains than children or adolescents. The most evidenced adult interventions are: sustained aerobic exercise (small-to-moderate effect on executive function and processing speed), optimising sleep quality to remove chronic deprivation (can recover several points of performance), and continued formal education or intellectually demanding learning. Most commercial brain training products do not produce measurable gains in general IQ in adults.
Does reading increase IQ?
Reading is associated with higher IQ, but the direction of causation is partially reversed: higher IQ people tend to read more. For children, extensive reading has a genuine causal effect on verbal IQ and knowledge-based reasoning — building vocabulary, world knowledge, and the ability to process complex syntax. For adults, reading builds crystallised intelligence (Gc) but has limited effect on fluid intelligence (Gf) — the pattern-recognition and reasoning component that IQ tests most emphasise. To improve Gf, sustained education and aerobic exercise have stronger evidence than reading alone.
How long does it take to improve IQ?
There is no single timeline. Sleep optimisation can improve cognitive performance within days of correcting chronic deprivation. Exercise effects on cognitive performance emerge after 4–8 weeks of regular aerobic training. Education shows IQ gains of 1–5 points per year of schooling accumulated over years. Nutritional corrections in deficient populations show gains over months to years. Apps and supplements claiming dramatic IQ improvements within weeks are not supported by evidence.
Do brain training apps like Lumosity actually work?
Not for general IQ. A 2014 open letter signed by 75 cognitive scientists found no compelling evidence that commercial brain training improves general cognitive ability. Lumosity was fined $2 million by the FTC in 2016 for making deceptive claims. People improve at the specific trained tasks — but this does not transfer to untrained cognitive tasks or general IQ performance. The near-vs-far transfer problem means app-based training improves app scores, not intelligence.
What foods increase IQ?
In populations with nutritional deficiencies, correcting them produces significant gains: iodine supplementation in iodine-deficient populations can raise IQ by 10–15 points, and iron correction in anaemic children shows 4–5 point gains. In well-nourished populations, no specific food or supplement has shown reliable IQ gains in randomised controlled trials. Omega-3 fatty acids show modest effects in some studies but the evidence is inconsistent. Maintaining overall diet quality, avoiding neurotoxins (alcohol, chronic lead exposure), and preventing deficiencies is the nutritional evidence base.
Does exercise increase IQ?
Regular aerobic exercise improves several cognitive abilities that contribute to IQ test performance — particularly executive function, working memory, and processing speed — with small-to-moderate effect sizes in meta-analyses. The mechanism involves increased BDNF, improved cerebrovascular health, and reduced neuroinflammation. The evidence is strongest for children and older adults; effects in healthy young adults are present but smaller. Aim for 150–200 minutes of moderate-intensity aerobic exercise per week, with measurable cognitive effects typically emerging after 4–8 weeks.