Skip to content
IQ Centre

Can you improve your IQ score?

Yes and no. Your score on a particular test can go up, sometimes noticeably, through familiarity with the puzzles and better test conditions. Your underlying reasoning ability is much harder to shift, and most products that promise to raise it don't have good evidence behind them. This guide separates the two, and then gives practical strategies for each type of puzzle so you can show what you're actually capable of.

In short: practice with the puzzle types and good test conditions reliably add a few points to a score. Education raises scores over years. Brain-training games mostly make you better at the games. On the day, the biggest avoidable losses come from tiredness, distraction and a small screen.

Practice: real gains, with a catch

People almost always score a few points higher the second time they take a reasoning test, even a different version of it. Psychologists call this the practice effect. It happens because the first time, part of your effort goes into working out what kind of puzzle you're looking at and what kinds of rules are allowed. The second time, you already know. Matrices, for instance, nearly always use the same few rule types: things that stay the same, change step by step, or rotate through three values. Once you know that, you search for the answer much more efficiently.

That's a genuine improvement in your score, and arguably a fairer reflection of your ability than a first attempt spent partly confused. But it doesn't mean you've become more intelligent, and the gains level off quickly: the jump from first to second attempt is the biggest, and after a few attempts there's little more to gain. It's also why professional tests are given to people who haven't seen them before, and why a good online test gives you puzzles you haven't seen when you take it again.

Brain training

Brain-training apps are built on an appealing idea: exercise your memory and attention, and your general intelligence will grow, like a muscle. The research hasn't been kind to it. A large review published in 2016 by Daniel Simons and colleagues in Psychological Science in the Public Interest examined the evidence cited by brain-training companies and concluded that training improves performance on the trained tasks, and sometimes on closely similar ones, but found little evidence that it improves everyday cognitive performance or general reasoning. Studies of working memory training specifically, including a widely cited 2013 meta-analysis by Monica Melby-LervÄg and Charles Hulme, reached a similar conclusion: gains on memory tasks, but no reliable transfer to measures like matrix reasoning.

That doesn't make these games bad. If you enjoy them, play them. Just don't expect them to change your IQ score, except on puzzles that closely resemble the games.

Education and learning

The best evidence for a lasting effect on intelligence scores comes from education. A 2018 meta-analysis by Stuart Ritchie and Elliot Tucker-Drob, which pooled studies designed to separate cause from effect, estimated that each additional year of education raises IQ scores by roughly one to five points. Reading widely, learning new skills and doing things that make you think hard are the everyday versions of the same thing. The effects build over years, not weeks, but they're real.

Test conditions: the easiest points

Many people lose points on the day for reasons that have nothing to do with ability. These are the easiest to fix:

  • Sleep. Short or poor sleep measurably hurts attention and working memory, both of which every reasoning puzzle relies on. Don't take a test at the end of a long day if you can avoid it.
  • One thing at a time. Notifications and interruptions break the chain of reasoning on multi-step puzzles. Silence your phone and close other tabs.
  • A bigger screen. Pattern grids and 3D figures contain fine detail. A laptop or tablet makes them easier to read than a small phone.
  • Pen and paper. Writing down the gaps in a number sequence or the rules you've found in a grid frees up working memory for the reasoning itself. It's not cheating; it's how many people do these tests best.
  • Pace. If a test isn't timed, don't rush. Most mistakes on mid-level puzzles come from picking an answer that fits the first rule you found, before checking the rest.
  • Guess when stuck. On most tests, including ours, a wrong answer costs nothing more than no answer, so never leave one blank.

Strategies for each type of puzzle

These come straight from how people who do well on each type tend to work. You'll find worked examples of every type in IQ test questions explained.

Pattern grids

  • Decide what the missing piece must be before you look at the options.
  • Check one feature at a time: number, size, shading, shape.
  • A rule has to work in every row, not just the first.
  • For line patterns, look at a line that's in both of the first two squares: is it still there in the third?

Number sequences

  • Write the gaps between numbers underneath. Then the gaps between the gaps.
  • If numbers grow quickly, try multiplication.
  • If the sequence zigzags, look at every other number.

Mental rotation

  • Pick a landmark, such as the end of the longest arm, and find it in each option.
  • Describe the path from there as a series of turns, and compare.
  • If everything matches except the direction of one bend, it's the mirror image.

Paper folding

  • Count first: each fold doubles the holes.
  • Unfold in reverse, starting with the last fold.
  • Diagonal folds swap a hole's distance from the top for its distance from the side.

Verbal reasoning

  • For analogies, say the relationship as a sentence: "a glove covers a hand".
  • For odd one out, find the narrowest category that fits all but one.
  • For deduction, only accept what the statements guarantee, however likely something else seems.

Memory span

  • Group the sequence into small chunks, such as two or three blocks at a time.
  • Remember the path as a shape rather than single positions.

Should you retake a test?

If your first attempt was on a phone, late at night or full of interruptions, a second attempt in better conditions will give a fairer picture. Expect your score to rise a little anyway from familiarity, and judge the change against the margin of error shown in your report rather than the single number: a move from 112 to 116 is within the normal wobble of any short test.

IQ Centre members can take a new test whenever they like, built from puzzles they haven't seen, and see their scores over time. There are also practice sets for each puzzle type, with the answer explained after every one, which is the quickest way to learn the patterns.

Ready to try?

The IQ Centre test takes about 15 minutes.