Do video games raise IQ?

There is no evidence that they raise general intelligence. Action games have been linked to better visuospatial and attention task performance, but a meta-analysis that matched control-group quality found the training effect nearly disappears. Any gains are confined to abilities close to those the game uses.

Are video games bad for the brain?

There is no robust evidence that gaming itself lowers cognitive ability either. A 2022 study of about 2,000 children found that those who played three or more hours a day did better on an inhibition task and a working-memory task. It was cross-sectional, though, and any time taken from sleep or study is a separate issue.

Which kinds of games are linked to cognition?

The reported effects come mainly from action games (first-person shooters and similar titles that demand fast decisions and rapid shifts of attention). Puzzle and brain-training games improve performance on the game itself, but transfer to other tasks is limited.

Are Video Games Good or Bad for Your Brain? What the Meta-Analyses Show

Published: 2026-09-08

In brief: The effect of video games on cognition depends on the type of game and how it is measured. In a meta-analysis by Bediou and colleagues (2018, Psychological Bulletin), habitual action gamers outperformed non-gamers on cognitive tasks (g ≈ 0.55), and intervention studies that trained non-gamers found a moderate effect (g ≈ 0.34). Gains were largest for visuospatial cognition and attention and small for inhibition and verbal tasks. Sala and colleagues (2018, same journal) reported that training effects nearly vanish once control groups are matched, and flagged publication bias. There is no evidence that games raise general intelligence (g); any effect is confined to domains close to the trained task.

"Too much gaming makes you dumb" and "games train reflexes and judgment" are both claims you hear all the time. Follow the research and the answer turns out to depend on the type of game and on what you measure. This article lines up the key studies on each side and separates what can be said today from what cannot.

What the action-game research showed

The starting point of the field is Green and Bavelier (2003, Nature). Habitual action gamers did better on visual attention tasks, and when non-gamers were trained on an action game for 10 days (one hour a day), their performance on the same tasks improved. The study matters because it showed that "gamers were already good at this" cannot be the whole explanation.

The later studies were pooled by Bediou and colleagues (2018, Psychological Bulletin). Their conclusion has two parts.

Gamers versus non-gamers (cross-sectional): gamers score higher on cognitive tasks, with an effect size of g ≈ 0.55, moderate to large by psychology's standards
Intervention studies training non-gamers: the effect shrinks to g ≈ 0.34, small but not zero

The largest gains were in visuospatial cognition (rotation and locating objects) and top-down control of attention. For inhibition, multitasking, and verbal tasks the effects were small or unclear. In other words, what improves is close to the abilities an action game actually uses.

The rebuttal in the same journal

The same year, the same journal published the opposite conclusion. Sala, Tatlidil, and Gobet (2018, Psychological Bulletin) reanalyzed whether game training improves cognitive ability and reported that the overall effect is small and shrinks as study quality rises.

The key issue is the control group. Comparing a trained group with a do-nothing group adds the effects of expectation and of any novel activity to the training. Comparing it with a group that did some other activity (an active control) subtracts those. In Sala's analysis, studies with active controls showed effects close to zero. They also found signs of publication bias, with smaller studies reporting larger effects.

The two meta-analyses differ in scope (action games only versus games in general) and in how studies were selected. Rather than asking which is right, the sensible reading is that the narrow claim about action games survives while the broad claim that games in general improve cognition does not. The same pattern recurs in brain-training hobbies and bilingualism and executive function.

Data on about 2,000 children

What about the claim that games harm the brain? Analyzing participants in the large ABCD brain-development study in the United States, Chaarani and colleagues (2022, JAMA Network Open) compared about 2,000 children aged 9 to 10, splitting them into those who never played video games and those who played three or more hours a day.

The gamers did better on a stop-signal task (inhibition) and an n-back task (working memory), and brain imaging showed higher activity in regions tied to attention and memory. But the study is cross-sectional. It cannot tell whether gaming improved these abilities or whether children who are already good at such tasks are drawn to games, a point the authors themselves state.

The study also measured cognitive task performance, not grades or sleep. If gaming displaces sleep, exercise, or study time, those effects arise separately. As discussed in the digital dementia article, evidence that screen time itself damages the brain is weak, but what it replaces matters.

How to read it: effects differ by domain

Putting this together, the effect of games is not a single good-or-bad axis. It has to be read domain by domain.

Two people can share the same overall IQ while having completely different profiles — the shape of the scatter often matters more than the single number.

Visuospatial cognition and visual attention: action gamers score higher and training has some effect. This is the best-supported domain
Processing speed and switching: effects are reported, but results change with control-group quality
Working memory and inhibition: linked in cross-sectional studies, but training effects are unclear
Verbal ability and general intelligence (g): no robust evidence of an effect

The general rule of training research is that the trained task improves, similar tasks benefit, and transfer to intelligence in general (far transfer) is limited. Games are no exception. The principle is covered in detail in Can you raise your IQ?.

Gaming experience shows up in domain scores

The domains most tied to gaming are spatial reasoning, which handles rotation and relative position, and pattern recognition, which is about spotting regularities quickly. How each domain relates to overall IQ is summarized in our skill correlation data.

If you want to check your strong domains, BrainRank's free IQ test scores 20 questions (about 10 minutes) across four domains, spatial reasoning, pattern recognition, logical reasoning, and classification, using item response theory (IRT), and returns your profile by domain. It is a good way to see where abilities built through gaming show up. Results are estimates for entertainment and self-understanding, not a medical or diagnostic assessment.

Frequently asked questions

Do video games raise IQ?: There is no evidence that they raise general intelligence. Action games have been linked to better visuospatial and attention task performance, but a meta-analysis that matched control-group quality found the training effect nearly disappears. Any gains are confined to abilities close to those the game uses.
Are video games bad for the brain?: There is no robust evidence that gaming itself lowers cognitive ability either. A 2022 study of about 2,000 children found that those who played three or more hours a day did better on an inhibition task and a working-memory task. It was cross-sectional, though, and any time taken from sleep or study is a separate issue.
Which kinds of games are linked to cognition?: The reported effects come mainly from action games (first-person shooters and similar titles that demand fast decisions and rapid shifts of attention). Puzzle and brain-training games improve performance on the game itself, but transfer to other tasks is limited.

Share this article

Editorial note & disclaimer

BrainRank Editorial Team

This article was written and edited by the BrainRank Editorial Team with reference to academic literature on psychometrics, including CHC theory and Item Response Theory (IRT). Statistics and percentages are calculated from a normal distribution model with a mean of 100 and a standard deviation of 15.

The tests on this site provide estimates for entertainment and self-understanding purposes only. They are not medical or clinical assessments, nor official psychological (intelligence) tests.

Sourcing standards and correction policy →