How to Increase Your IQ - Brain Training That Works

What works vs what does not - the evidence table

Here is the whole field at a glance, ranked by how strong the evidence is and how large a realistic effect you should expect. "Effect" here means change in measured cognitive performance, not a guaranteed permanent IQ jump.

Intervention	Evidence strength	Realistic effect	The honest catch
Staying in education	Strong (large meta-analysis)	~1-5 IQ points per extra year	Effect is real and durable, but you have to actually do the schooling
Aerobic exercise	Moderate	Small-to-moderate on executive function (g ≈ 0.13-0.16)	Benefits fade if you stop; not a one-off fix
Adequate sleep (7-9h)	Moderate-strong	Restores attention, memory, reasoning	Fixes a deficit rather than pushing above baseline
Correcting a nutrient deficiency	Moderate (deficiency only)	Meaningful if you were deficient	Supplements do little if you are already well-nourished
Learning a hard new skill	Suggestive	Domain-specific gains, weak transfer	You get better at the thing, not at "everything"
Working-memory games (dual n-back)	Weak / contested	Near-zero transfer to general intelligence	Improves the game; robust transfer has not replicated
Commercial "brain-training" apps	Weak	Practice effect only	You are paying to get good at their puzzles

The pattern is blunt: the top of the table is lifestyle and learning, the bottom is the stuff sold as a shortcut.

Education - the strongest lever by a wide margin

If you do only one thing on this list, stay in education longer. The landmark evidence is a 2018 meta-analysis by Stuart Ritchie and Elliot Tucker-Drob in Psychological Science, which pooled 142 effect sizes from 42 datasets covering more than 600,000 people. Using three types of quasi-experimental design, controlling for earlier intelligence, exploiting policy changes to school-leaving age, and comparing children either side of school-entry cutoffs, they converged on the same answer: each additional year of education is associated with a gain of roughly 1 to 5 IQ points.

Two details make this the gold standard. First, the effect showed up across every broad category of cognitive ability they tested, not just the school-like ones. Second, it appeared to persist across the lifespan rather than washing out. The authors called education the most consistent, robust, and durable method yet identified for raising intelligence. "Education" here is broad: formal study, but also the deep, effortful, sustained learning that schooling forces on you. The takeaway for an adult is not to re-enroll in high school; it is that structured, demanding learning over years, a degree, a trade qualification, a serious language, moves the needle in a way no app has ever matched.

Lifestyle - exercise, sleep, and nutrition

These will not turn an average score into a genius one, but they remove the drags that quietly cost you points, and unlike apps they are backed by convergent evidence.

Aerobic exercise

Regular aerobic activity, the kind that raises your heart rate, has moderate support for improving executive function, the planning, focus, and mental-flexibility skills that overlap with reasoning tests. Meta-analyses of healthy adults typically find small-to-moderate benefits, with effect sizes in the region of g ≈ 0.13 to 0.16 for executive function, plus measurable gains in working memory. One proposed mechanism is brain-derived neurotrophic factor, or BDNF, a protein that supports the growth and survival of neurons and rises with exercise. The realistic expectation: exercise is a genuine cognitive tune-up, not a transformation, and the benefit depends on keeping it up.

Sleep

Sleep is the lever most people ignore while chasing apps. The evidence that sleep loss degrades attention, working memory, and reasoning is strong, and adults broadly do best on 7 to 9 hours. The important framing: fixing your sleep does not push you above your ceiling, it stops chronic sleep debt from dragging you well below it. If you are running on 5 hours, no training program will show its true effect, because you are testing a sleep-deprived version of yourself.

Nutrition

Nutrition matters mostly at the deficiency end, and this is where honest reporting departs from supplement marketing. For omega-3 fatty acids the picture is split: a meta-analysis of randomized trials in children, adolescents, and young adults found no overall effect of omega-3 supplementation on cognitive test performance, yet observational work links low omega-3 status and low fish intake to poorer cognition and lower measured IQ. The reconciliation is straightforward: correcting a genuine deficiency helps, topping up someone already well-nourished does little. Treat food, including oily fish, iron, and iodine, as insurance against deficits rather than a nootropic.

Lifestyle lever	Practical target	What to realistically expect
Aerobic exercise	Most days, enough to raise heart rate	Sharper executive function; fades if you stop
Sleep	7-9 hours, consistent schedule	Removes a large hidden drag on reasoning
Nutrition	Fix deficiencies; oily fish, iron, iodine	Helps if deficient; little extra above sufficiency

The brain-games problem - working memory and the transfer trap

Now the part the industry does not advertise. The whole "train your brain" category rests largely on one famous study: Jaeggi and colleagues in 2008, in PNAS, reported that healthy young adults who trained on the dual n-back task, a demanding working-memory exercise, improved on tests of fluid intelligence, with a dose-response pattern in which more training meant bigger gains. It was an exciting result and it launched an industry.

Then replication happened, and the result did not hold up well. A key weakness of the original was that the comparison group did no training at all, a passive control. When later studies used active control groups, people doing some other engaging task, the transfer to general intelligence largely vanished. Broader reviews reached the same conclusion: working-memory training reliably makes you better at the trained task and very similar tasks, but it does not robustly raise underlying fluid intelligence. Part of the original effect is now attributed to motivation, a rested passive-control group simply trying less hard on the post-test than the trained group.

This is the transfer problem in one paragraph. Getting good at n-back makes you good at n-back. Getting good at a matching game makes you good at that matching game. The skill does not spill over into the general reasoning that IQ tests measure. So if your goal is a higher underlying ability rather than a higher score on one familiar puzzle, commercial brain-training apps are close to the weakest option on this page.

Realistic expectations - what to actually do

Put the evidence together and the plan is simple, if not easy:

1. Learn something hard, for years. This is the one lever with large, durable, replicated support. Pick a real qualification or skill, not a quiz.
2. Move your body most days. Aerobic exercise gives a modest, real edge and protects the brain long-term.
3. Sleep 7 to 9 hours. Stop paying a nightly tax on your own cognition.
4. Fix nutritional gaps, do not chase supplements. Insurance against deficiency, not a shortcut.
5. Skip the brain-training apps as an intelligence tool. They are fine as entertainment; they are not a transfer machine.

And when you take an IQ test, expect a practice effect on the second sitting, that bump is the format, not new brainpower. If you want a clean read on where you actually stand, take a properly normed test once, rather than grinding the same one repeatedly.

Frequently asked questions

Q: Can you actually increase your IQ?

A: Partly, and honestly. Adult IQ is 50 to 80 percent heritable and fairly stable, so large permanent jumps are not realistic. But staying in education raises measured intelligence by roughly 1 to 5 points per additional year, and exercise, sleep, and fixing nutritional deficiencies remove drags that quietly cost you performance. You are closing the gap to your ceiling, not rewriting your genes.

Q: Do brain-training apps and games work?

A: Not for raising general intelligence. The original 2008 dual n-back study did not replicate well once researchers used proper active control groups. The consistent finding is that these games improve the trained task but do not transfer to the broad reasoning IQ tests measure. Enjoy them as puzzles, not as an intelligence upgrade.

Q: What is the single most effective way to improve intelligence?

A: Education. A 2018 meta-analysis of more than 600,000 people found each extra year of schooling was linked to about 1 to 5 IQ points, across all cognitive domains and persisting across life. No other intervention has evidence of that scale, consistency, or durability. Sustained, demanding learning is the closest thing to a proven method.

Q: Does exercise really make you smarter?

A: It gives a modest, real boost, especially to executive function. Meta-analyses of healthy adults find small-to-moderate gains, with effect sizes around g ≈ 0.13 to 0.16, likely linked in part to exercise-driven BDNF. It is a genuine tune-up rather than a transformation, and the benefit depends on keeping the habit going.

Q: If I score higher the second time I take an IQ test, did my IQ go up?

A: Usually no, that is a practice effect. Repeating the same or similar test makes you better at that specific format without changing your underlying ability. To gauge real change, use a properly normed test and do not grind the same one over and over.

References

• Ritchie, S. J., & Tucker-Drob, E. M. (2018). How Much Does Education Improve Intelligence? A Meta-Analysis. Psychological Science, 29(8), 1358-1369. https://journals.sagepub.com/doi/abs/10.1177/0956797618774253
• Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. PNAS, 105(19), 6829-6833. https://www.pnas.org/doi/10.1073/pnas.0801268105
• Bratsberg, B., & Rogeberg, O. (2018). Flynn effect and its reversal are both environmentally caused. PNAS, 115(26), 6674-6678. https://www.pnas.org/doi/10.1073/pnas.1718793115
• American Psychological Association - Intelligence and cognitive research overview. https://www.apa.org/topics/intelligence

Last updated: July 13, 2026