"Gamers have faster reflexes" sounds obvious. The research is more interesting, and less settled, than that. Habitual action game players do score better on many lab tasks. Whether the games caused that difference, and whether it carries over to anything outside games, is still argued over by the people who study it. Here's what the evidence supports, and how to check your own reaction time without fooling yourself.
What researchers mean by "action video games"
Most of this research isn't about video games in general. It focuses on action video games: fast-paced games, usually first- or third-person shooters, that demand quick responses, tracking several things at once, and switching between narrow and broad attention. Training studies usually compare them with control games that are slower or less demanding. In the training studies reviewed by Bediou and colleagues (2018), control games included The Sims, puzzle games and Tetris.
That distinction matters. A finding about action games doesn't automatically apply to strategy, puzzle or mobile games.
The finding that started it
In 2003, Green and Bavelier published a study in Nature comparing habitual action game players with non-players on visual attention tasks. The players did better across several experiments. In a further experiment, non-players trained on an action game improved compared with their own pre-training scores. That training result is what turned a correlation into a causal claim.
In a 2009 review, Dye, Green and Bavelier went further. They argued that action game experience is linked to faster responses across a range of perceptual tasks without the usual loss of accuracy, and suggested that action games could act as training for faster perceptual reaction times. At the time, they could point to only a few training studies showing that faster responses could be trained by game play.
What the meta-analyses say
Since then, enough studies have accumulated to pool them. The results depend heavily on the type of study and how strictly publication bias is handled:
| Analysis | What it pooled | Headline result | Main caveat |
|---|---|---|---|
| Bediou et al. (2018) | Cross-sectional: players vs non-players | Players better by about half a standard deviation (g = 0.55) | Can't show cause and effect |
| Bediou et al. (2018) | Training studies with control games | Smaller effect, about a third of a standard deviation (g = 0.34) | Authors flagged publication bias as a threat |
| Hilgard et al. (2019) | Re-analysis of the 2018 training studies | Minimal evidence of transfer after stronger bias adjustments | Also found overlapping participant samples reported across papers |
| Sala et al. (2018) | Video games broadly, three meta-analyses | Small or null effects; no evidence of a causal benefit to cognitive ability | Broader set of games and outcomes |
| Bediou et al. (2023) | 28 training studies with active controls | Small effect (g = 0.30, 95% CI 0.11 to 0.50) | Notable heterogeneity in a small set of studies |
Two things stand out. First, these effect sizes are mostly pooled across many cognitive measures, such as attention, spatial cognition and perception. They are not a pooled estimate of simple reaction time in milliseconds. Second, the strongest effect comes from the design least able to show cause. Cross-sectional comparisons show bigger differences than randomized training studies.
Correlation is not causation
A difference between gamers and non-gamers can come from several places:
- Self-selection. People who are already quick may enjoy action games more, do well at them, and keep playing. The games didn't make them fast; being fast made the games rewarding.
- Expectations. Boot, Blakely and Simons (2011) pointed out that studies often recruited through ads seeking "expert" gamers, which can tell participants how they're expected to perform. Motivated participants try harder on game-like lab tasks.
- Other differences. Age, sleep, screen habits and comfort with computers can all differ between groups.
Training studies get around the direction problem by assigning people to games at random, but they have their own weak points: control groups that don't match expectations, small samples, and the publication-bias concerns in the table above.
Transfer: you mostly get better at what you practice
Getting better at something you didn't directly practice, called far transfer, is hard to show in general. Sala and colleagues concluded that video game training is no exception. Practice reliably improves the practiced game. Gains on unrelated tasks are much less certain.
Simple reaction time also has a narrow range to work in. One cue, one response and nothing to predict leaves little room for strategy. Much of what makes experienced players look fast is anticipation and better decisions, the same pattern seen in reaction time in sports. Those skills matter, but they show up in choice tasks, not in a simple-reaction baseline. See simple vs choice reaction time for the difference.
How to test your own change honestly
You can't run a randomized trial on yourself, but you can avoid the most common mistakes. Use a phased plan:
| Phase | Weeks | What you do | What you record |
|---|---|---|---|
| Baseline | 1–2 | Your usual routine, no new game | Several sessions a week at the same time of day |
| Change | 3–6 | Add the action gaming you want to test | The same sessions, same setup |
| Return | 7–8 | Go back to your baseline routine | The same sessions, same setup |
Rules that keep the comparison fair:
- Fix the setup. Use the same device, display refresh rate, input, browser and seat. A hardware change will mix with any training effect.
- Test before playing, not after. A long session can leave you warmed up one day and tired the next.
- Record medians and spread, not best scores. Read how many trials you need before trusting a small difference.
- Expect practice effects. Doing the test over and over is training too. If your scores keep improving through the return phase, practice on the test is a likely explanation.
- Accept "inconclusive". If the phase difference is within your normal session-to-session spread, you haven't shown a change in either direction.
You'll still be unblinded, since you know what you're testing, so treat the result as evidence about you on your setup, not proof that games work.
Sources & context
Original studies: Green and Bavelier, Action video game modifies visual selective attention (Nature, 2003), and Dye, Green and Bavelier, Increasing speed of processing with action video games (Current Directions in Psychological Science, 2009). Meta-analyses and critiques: Bediou et al., Meta-analysis of action video game impact on perceptual, attentional, and cognitive skills (Psychological Bulletin, 2018); Sala, Tatlidil and Gobet, Video game training does not enhance cognitive ability (Psychological Bulletin, 2018); Hilgard, Sala, Boot and Simons, Overestimation of action-game training effects (Collabra: Psychology, 2019); Bediou et al., Effects of action video game play on cognitive skills (Technology, Mind, and Behavior, 2023); and Boot, Blakely and Simons, Do action video games improve perception and cognition? (Frontiers in Psychology, 2011). These studies use lab tasks and composite cognitive measures, not WiredReflex, so none of them predicts a specific millisecond change in your score. WiredReflex is a browser-observed timing tool, not a cognitive or medical assessment. See how we measure.
Frequently asked questions
+Do video games improve reaction time?
Possibly a little, for some lab tasks, but the evidence is contested. Training studies of action video games report small average gains across a range of cognitive measures, and re-analyses that correct more strongly for publication bias find little evidence of transfer. No study shows a fixed millisecond improvement you can count on.
+Do gamers have faster reaction times than non-gamers?
On average, habitual action game players do better than non-players on many lab tasks; a 2018 meta-analysis put the cross-sectional difference at about half a standard deviation. That comparison can't show that the games caused the difference, because people who are already fast may be more likely to play and stick with action games.
+What kind of video games are linked to faster reactions?
The research focuses on action video games: fast-paced games, often first- or third-person shooters, that demand quick responses and divided attention. Training studies usually compare them with control games such as puzzle or life-simulation games.
+How can I tell if gaming improved my reaction time?
Test on the same device, at the same time of day, several times a week for at least two weeks before you change anything, then keep the same routine during and after the change. Compare medians and spread across phases, not best scores, and remember that getting better at the test itself is also practice.