Gaming and Mental Health: What the Research Actually Says
Updated August 5, 2026
In 2020, a pandemic supply shortage meant Japanese retailers could not meet demand for the Nintendo Switch and PlayStation 5, so they handed out consoles by lottery. Without meaning to, they ran the experiment that gaming research had been missing for thirty years: a random assignment of who got a games console and who did not.
Researchers found the winners. The result, published in Nature Human Behaviour in 2024, is the strongest causal evidence available on whether playing games is good for you.
The research on games and mental health measures two separate things: whether having access to games at all changes how you feel, and whether the number of hours you play changes it. Only the first shows a clear effect, and most of the confusion in this field comes from articles treating the two questions as one.
Why is a console lottery such a big deal?
Because almost all research before it was correlational, and correlation cannot tell you which way the arrow points. If unhappy people play more games, a survey cannot distinguish “gaming causes distress” from “distress causes gaming” from “a third thing causes both.”
A lottery breaks that deadlock. Who wins is random, so the winners and losers differ in nothing except the console. Any difference in wellbeing afterwards has one plausible source.
Egami and colleagues analyzed a survey of 97,602 people covering 2020 to 2022, with 8,192 lottery participants forming the causal core of the study.
What they found: console ownership reduced psychological distress and improved life satisfaction. Winning a Switch lottery cut psychological distress by roughly 0.2 standard deviations; a PS5 win cut it by about 0.1 and raised life satisfaction by around 0.2. Adjusting for the fact that not everyone who won actually played, the estimated effect of owning and using a console ran considerably higher, up to 0.6 standard deviations for the Switch.
These are meaningful but moderate effects. Nobody is claiming a games console treats clinical depression.
So why did Oxford find nothing?
Because it asked a different question. Both results stand.
The Oxford Internet Institute ran a study with something almost nobody else has: real playtime telemetry. The team partnered with seven publishers, Nintendo, EA, CCP, Microsoft, Sony, Square Enix and Ubisoft, to get six weeks of objective play data from 38,935 players, so nobody had to estimate their own hours. It found little to no evidence of a causal connection between how much someone played and how they felt.
The distinction is exact and worth holding onto. The Japanese study measured whether you have access to games at all. The Oxford study measured how many hours you put in among people who already play. Getting a console when you had none changes something. Playing 14 hours a week instead of 10 apparently does not.
Oxford’s more striking finding was about motive. Players who felt they had to play reported worse wellbeing than those who played because they wanted to, regardless of hours, and the 2022 paper found the pattern held over time: play experienced as a voluntary choice predicted better wellbeing later on.
Hours are the wrong variable. That is the single most reproducible result in this field, and it is the one every “screen time limit” headline gets wrong.
Is there a point where more gaming stops helping?
The Japanese data suggests yes, and puts it in a specific place: the psychological benefits diminished once daily play exceeded about three hours.
That is a curve flattening out, and it comes from one study in one country during a pandemic. But it fits the shape of the other evidence: a moderate amount of something you chose to do is good for you, and the benefit levels off as the amount climbs without ever turning negative.
Does the type of game matter?
Almost certainly, and this is where the sourcing falls apart.
You will find confident claims that cozy and simulation games produce the largest anxiety reductions of any genre. We went looking for the study behind that and could not find one. The design mechanisms that make those games feel calming are real enough, and we have written about them in why cozy games are so relaxing, but a felt effect is not a measured outcome. There is no robust head-to-head comparison of genres against mental health outcomes in the peer-reviewed literature we could locate. If such a study exists, it is not the one being cited. The audience for those games turns out to be just as unmeasured, which we went through in who actually plays cozy games.
What does exist is more specific and more interesting. A 2025 study in JMIR Serious Games, a survey of 2,818 Chinese college students playing during pandemic lockdowns, concluded that the buffering effect may be strongest among routine players with moderate to low anxiety and depression symptoms, while heavy play among people already struggling with high symptoms may have detrimental consequences. It is a cross-sectional survey, so it cannot establish direction. Still, the same activity appeared to run different ways depending on who was doing it.
A separate 2025 randomized controlled trial in the same journal took 1,105 users of the MMO Pigg Party and tested a behavioral activation intervention delivered inside the game, finding a small improvement in depressive symptoms (SMD 0.12) and no improvement at all in social anxiety. Among participants with a diagnosis of depression, the effect did not hold. That is a game used deliberately as a delivery vehicle for a therapeutic technique, which is a different claim from “playing games makes you feel better.”
Four studies carry most of the weight in this piece, and they are easier to keep straight side by side:
| Study | Design and sample | What it found |
|---|---|---|
| Egami et al., Nature Human Behaviour, 2024 | Japanese console lotteries; survey of 97,602 people, 8,192 lottery participants | Console ownership reduced psychological distress by 0.1 to 0.6 standard deviations; a PS5 win also raised life satisfaction |
| Oxford Internet Institute, 2022 | Six weeks of publisher playtime telemetry from 38,935 players across seven publishers | Little to no causal link between hours played and wellbeing; felt compulsion predicted worse wellbeing |
| JMIR Serious Games, 2025 | Cross-sectional survey of 2,818 Chinese college students during lockdowns | Buffering effect strongest for routine players with moderate to low symptoms; possible harm at high symptoms |
| JMIR Serious Games, 2025 | Randomized controlled trial, 1,105 users of the MMO Pigg Party | Small gain on depressive symptoms (SMD 0.12), none on social anxiety, none among diagnosed participants |
No two of them measure the same thing, which is how they can all be right at once.
So the defensible summary: who you are and why you are playing predict the outcome better than what you are playing. Genre probably matters. It has not been measured well enough to quantify.
A note on the specific percentages you will see quoted
Claims like “moderate gaming reduces depression symptoms by 17%” circulate constantly, usually with a real journal name attached to give them weight. We could not source that figure, or several like it.
The effect sizes in the actual literature are reported in standard deviations, are moderate, and come hedged with conditions. Any article giving you a clean percentage for a whole population is either citing one narrow study far outside its scope or citing nothing at all.
How should you read this for your own life?
Four things follow from the evidence, and none of them is a rule about hours.
- Ask whether you want to play or feel you have to. That distinction is what Oxford’s data picked up over and over. Duration barely registered.
- Notice whether it is a break or an avoidance. The JMIR lockdown result suggests the same session can help someone at moderate baseline and hurt someone already struggling.
- Three hours a day is roughly where the returns flatten, on the best evidence available. Nobody has drawn a line past it.
- Absorption is not the same as compulsion. Losing track of time in a game is a sign the difficulty matched your attention, something we covered in flow state in games. The question to ask is whether you can stop when you want to.
None of that tells you which game to open tonight. If the honest answer is that work has flattened you, we sorted ten of them by the kind of exhaustion they suit in games to play when you are burned out.
Where our own game sits in this
We are Overbaked Studio, a small independent studio in Istanbul, making Grandma Left Me a Building, a cozy apartment management sim coming to Steam. We are a games studio writing about whether games are good for you, which is a conflict of interest we would rather name than pretend away.
So we will avoid the claim the genre invites. We are not going to tell you a cozy game improves your mental health, because the research supporting that specific claim does not exist yet. What the evidence does support is narrower: games people choose freely, that do not demand more than they want to give, tend to sit on the better side of the findings. Ours has no time pressure, so a session ends when you decide it does, and the 20+ building upgrades sit waiting whenever you come back, with nothing expiring while you are away. What we want pulling you back is curiosity about whether the eight-year-old tenant reassembles the thing he took apart, which he manages about 20% of the time, and not a sense of obligation. That is a design intention, not a health outcome, and with the game unreleased we have no evidence it works.
If you are struggling, a game is not a treatment, and the research above is careful about this even when the headlines are not.
Frequently asked questions
Are video games good for your mental health?
The best causal evidence, a 2024 Nature Human Behaviour study using Japanese console lotteries, found that owning a console reduced psychological distress by moderate amounts, and that PS5 owners also reported higher life satisfaction. The effects are real but modest, and gaming is not a treatment for clinical conditions.
How many hours of gaming per day is healthy?
The Japanese study found psychological benefits diminished beyond about three hours of daily play. Oxford’s telemetry research found that hours barely predicted wellbeing at all among existing players, and that motivation mattered far more.
Are cozy games better for anxiety than other genres?
Possibly, but no reliable study has compared genres head to head on mental health outcomes. Claims that cozy or simulation games top such a ranking are not traceable to published research.
When does gaming become a problem?
The research points at compulsion. Oxford found that players who felt obliged to play reported worse wellbeing regardless of hours, while the average effect of time spent playing on wellbeing was probably very small.
Key takeaways
- A console lottery in Japan produced the strongest causal evidence yet: across 97,602 surveyed people, console ownership reduced psychological distress by 0.1 to 0.6 standard deviations, and a PS5 win also raised life satisfaction.
- Oxford’s telemetry study of 38,935 players found little to no causal link between playtime and wellbeing, which does not contradict the above; access and hours are different variables.
- Motivation beats duration. Playing because you feel you have to is the pattern associated with worse outcomes.
- Benefits flatten past roughly three hours a day in the Japanese data, where the curve bends gradually.
- Genre effects are unmeasured. The claim that cozy games lead the field on anxiety reduction has no study behind it that we could find.
If you want a game built to stay on the “want to” side of that research, you can wishlist Grandma Left Me a Building on Steam.
