Online Games for Stress Relief: What Actually Works

Calm, absorbing games do help with stress, and there's real research behind it. But the effect is smaller and more conditional than the numbers you'll see quoted, and the biggest predictor is not the game you pick or the hours you put in. It's whether the session actually fits what you needed.

That's a less exciting answer than "Bejeweled cuts stress 54 percent." It's also the one the strongest evidence supports. Here's what holds up, what doesn't, and how to use a short session as a genuine reset.

Do games actually reduce stress?

Yes, for many people the right games reduce stress by absorbing your attention completely and giving your mind a break from worry. Psychologists call this state flow: being fully engaged in something challenging but achievable. Flow crowds out rumination, and short casual gaming sessions are a reliable way to reach it.

But the effect depends heavily on the game and your mindset. A relaxing puzzle can calm you, while a rage-inducing level can do the opposite. So the choice of game matters more than gaming itself, which is why this article keeps coming back to picking the right one.

What does the strongest research actually show?

The best evidence on this comes from the Oxford Internet Institute, because they did something almost nobody else has managed. They got game companies to hand over actual play data, so nobody had to guess how long they'd played.

Johannes, Vuorre, Magnusson and Przybylski studied 39,000 players across seven video game companies, combining objective telemetry with wellbeing surveys over six weeks. The games included Animal Crossing: New Horizons, Apex Legends, Eve Online, Forza Horizon 4, Gran Turismo Sport and The Crew 2. Their finding, published in Royal Society Open Science in July 2022: little to no evidence for a causal connection between how long people played and how they felt. What did matter was motivation.

A 2025 follow-up sharpened that. Ballou, Vuorre, Hakman, Magnusson and Przybylski tracked 703 casual adult Nintendo Switch players across 150 games and more than 140,000 hours of play. As Oxford reported, playtime had negligible correlation with wellbeing. What predicted it was whether players felt gaming was genuinely valuable in their lives, a measure the researchers called gaming life fit. That study was funded by the Huo Family Foundation and the UK Economic and Social Research Council, not by a games company.

There's now a review that pulls the trial evidence together, and it's the single most useful source on this question. Wols, Pingel, Lichtwarck-Aschoff and Granic went through 145 randomised controlled studies covering people aged 6 to 24, publishing in Clinical Psychology Review in March 2024. Seventy-five of those studies looked at clinical samples and seventy at healthy young people.

Their finding is more encouraging than the Oxford telemetry work, and the two aren't in conflict. Casual games came out most effective for depression, anxiety and ADHD, while purpose-built applied games did better on social skills and verbal memory. So short casual sessions do appear to shift how you feel. What Oxford adds is that racking up hours is not the mechanism.

But read the authors' own caveat. They called for clearer evaluation guidelines and standardised outcome measures, and the abstract reports no pooled effect size at all. A review of 145 trials that still can't give you one number is telling you something real about how varied this literature is. Anyone quoting you a precise percentage is not drawing it from here.

None of this says games don't help. It says the mechanism is attention and fit, not dose. A ten-minute round of solitaire that genuinely takes your mind off a bad afternoon is doing real work. Two hours of grinding something you've stopped enjoying is not, no matter what the genre label says.

Why should you be sceptical of the big percentages?

Because of who paid for them, and how few people were in the room.

You'll see two figures quoted constantly in articles on this topic. That casual games reduced physical stress by 54 percent, and that they cut depression symptoms by 57 percent. Both come from research at East Carolina University led by Carmen Russoniello, and both are real published findings rather than invented ones.

Two things almost never get mentioned alongside them.

• The games company funded the research. PopCap Games underwrote the studies, and every game tested (Bejeweled 2, Peggle and Bookworm Adventures) was a PopCap title. That doesn't make the results wrong, but it's a conflict of interest that belongs next to the number.
• The samples were tiny. The depression study randomised 30 people into the experimental group and 29 into the control. Around sixty people total, all meeting criteria for clinical depression. That's a pilot, not a settled result.

Compare that with 39,000 players and objective telemetry, and you can see why the smaller, louder numbers travel further. Dramatic percentages make better headlines than "motivation matters more than duration."

So use the casual-games research as a reason to try a calm game when you're tense. Don't use it as evidence that a specific game will cut your stress by a specific amount.

Which games lower stress and which raise it?

Stress-lowering games tend to be gentle, absorbing, and free of punishing pressure. Think solitaire, mahjong, calm puzzles, and casual arcade games you play at your own pace. They give you steady progress and a soothing rhythm without high stakes.

Stress-raising games are the opposite: intensely competitive, heavily time-pressured, or punishing, where losing feels costly. Reviews of the research note that high-stakes competitive play and long late-night sessions can activate or sustain stress biology rather than calm it. So if a game leaves you tenser than when you started, it is not the right choice for unwinding, no matter how popular it is.

That split is a decent rule of thumb, but one experiment is worth knowing about because it muddies it in a useful way. Porter and Goolkasian put 148 people through a two-by-two design for Frontiers in Psychology (2019, vol. 10, article 967), 37 per group, crossing how the task was framed against which game they played: Mortal Kombat or Tetris.

The bodies behaved as you'd guess. Mortal Kombat players showed lower heart rate variability while playing and higher blood pressure afterwards. Tetris players showed neither. But the self-reports went the other way. The fighting game players reported more positive emotion and less frustration than the puzzle game players did.

So "did that calm me down" has two answers depending on who you ask, your nervous system or you. Which is worth holding onto, because the advice below leans on how you feel afterwards, and this is a reminder that your read on it isn't a perfect instrument. If a fast, loud game genuinely leaves you in a better mood, that's real and it counts. Just don't assume it also left your body relaxed, especially close to bedtime.

Which genres has the research actually tested?

Fair challenge to the section above, since "calm games good, intense games bad" is exactly the kind of tidy claim this article keeps telling you to distrust. Somebody has checked, and the answer is messier than the tidy version.

Pallavicini, Pepe and Mantovani at the University of Milano Bicocca reviewed the evidence on ordinary commercial games, the ones you'd actually play rather than purpose-built therapy software, publishing in JMIR Mental Health in August 2021. They found 28 studies meeting their criteria, published between 2006 and 2021. Exergames were the most studied at 12 studies, casual games next at 9. Most of the work was on young adults, about 68 percent of studies, with older adults at 14 percent and children at 11 percent.

Two findings are worth your time.

The genre list is wider than you'd guess. Stress and anxiety reductions showed up not just for casual games and exergames but for action games, action-adventure games, and augmented reality games too. So the calm-puzzle stereotype isn't the whole story, and a game being fast or loud doesn't automatically disqualify it.

But action games cut both ways. The review found mixed results there: some studies showed action games raising stress, others showed them lowering it while also increasing happiness. Same genre, opposite outcomes, which is about as clear a signal as you'll get that the label on the box doesn't decide what happens to you.

The other finding backs up the advice further down this page. Play durations across the 28 studies ran from about 2 minutes to 15 hours, and the reviewers concluded that even single, short sessions had benefits for stress and anxiety. You don't need to clear an evening. A few minutes appears to be enough to register.

There's a second review worth knowing about, narrower and pointed straight at the kind of game you'd actually play in a browser tab. Russell Pine, Theresa Fleming, Simon McCallum and Kylie Sutcliffe looked only at casual videogames in the Games for Health Journal (2020). Thirteen studies met their criteria, covering nine different casual games, split across six studies on anxiety, four on stress or low mood, and two on depression.

Read their conclusion carefully, because the wording is doing real work. Casual games "may have promise" for anxiety, depression, stress and low mood. Not proven, not a treatment. Promise. Thirteen studies across nine games is a thin base to stand a claim on, and the authors don't pretend otherwise. That's the honest version, and it's roughly where the whole field sits: enough signal to justify opening a puzzle game when you're wound up, nowhere near enough to justify the numbers you see in headlines.

So the practical rule shifts slightly. Rather than "pick a calm genre," it's closer to "pick a game that absorbs you without punishing you, then check how you feel." For most people that lands on gentle puzzles anyway. But if a fast game genuinely relaxes you, the evidence isn't telling you you're wrong.

Why do absorbing games calm you down?

It comes down to attention and arousal. When a game absorbs you at just the right difficulty, your mind stops circling over problems and settles into the present. That mental break is the core of the stress-relief effect, and it is why a gentle puzzle can feel almost meditative.

There is a physical side too. Calm, enjoyable activity lets your body ease out of its stress response, while frustration and high stakes can trigger the release of stress hormones like cortisol.

Occupational psychology gives this a more precise shape than "it relaxes you," and the model is worth knowing because it tells you what to actually look for. Recovery research breaks the process into four separate experiences: psychological detachment (you stop thinking about the thing), relaxation (low effort, low arousal), mastery (you get better at something unrelated), and control (you decide what happens next). Games are unusual because they can deliver all four at once, where a TV show mostly manages the first two.

Koçak, Gorgievski and Bakker applied that framework to gaming specifically in Applied Psychology (2024, vol. 73, issue 3). They tested two separate routes from evening gaming to how people felt the next morning: a passive one running through detachment, and an active one running through mastery. So "did it help" splits into two questions. Did it get the day out of your head, and did it give you something to be good at.

The part that matters most for you is what they found alongside it. The same evening of gaming did different things depending on the player's relationship with the hobby, framed as harmonious versus obsessive passion. Playing because you want to, and being able to stop, points one way. Playing because you feel you have to points the other. Which is the honest answer to why the same game genuinely restores one person and leaves another flat.

The practical takeaway is that the feeling matters more than the label, so notice your own reaction rather than assuming any game relaxes you. If you enjoy the mental side of this, our guide to the benefits of puzzle games goes further.

Does using games to cope backfire?

It's the fair question to ask of an article like this one. If you reach for a game every time you feel stressed, are you building a coping habit or an avoidance one?

Someone tested it. Farmer and Lloyd ran a repeated-measures experiment on 40 university students, measuring stress before and after a gaming session, and published in Healthcare in 2024.

Two sets of results, and both are useful.

First, the session worked, and this is the concrete physiological evidence that's usually missing from articles on this topic. Pulse rate dropped by 4.7 milliseconds after playing, with a large effect size (d = 0.79, p < 0.001), measured by photoplethysmography rather than self-report. Positive mood rose by 3.32 points (d = 0.57, p < 0.001). Negative mood fell too, from a median of 13 to 11, with a substantial effect size (rB = 0.68, p < 0.001). So a short session measurably shifts both how you feel and what your body is doing.

Two small notes on precision, since this article keeps insisting on them elsewhere. The pulse result rests on 37 participants rather than the full 40, because not everyone yielded usable readings. And the paper reports a change in pulse rate in milliseconds, which is an odd pairing of measure and unit. That's how the study states it, so it's reproduced here rather than quietly tidied up, but it's the kind of thing worth noticing before you repeat a number.

Second, and this is the part worth knowing, they tested the worry directly. The theory they were checking is called compensatory use: the idea that the people who get the biggest mood boost from gaming are the same people most likely to slide into problematic use, because the relief is what keeps pulling them back.

It didn't hold up. There was no correlation between how much someone's mood improved and their gaming disorder score (ρ = -0.07, p = 0.62), and no association between pulse rate change and gaming disorder either (r = -0.15, p = 0.38). Getting real relief from a session did not mark someone as at risk.

The honest read is that using a short game to take the edge off a bad day looks like ordinary coping, not a warning sign. What that study can't tell you is where the line sits for you personally. The signals worth watching are the obvious ones: play that's crowding out sleep, work or people, or reaching for a game to avoid something rather than to recover from it. If either sounds familiar, the game isn't the problem to solve first.

There is an actual clinical line, though, and it's worth knowing where it sits rather than guessing. The World Health Organization added gaming disorder to the International Classification of Diseases, and its published criteria describe three things together: impaired control over gaming, giving gaming increasing priority "to the extent that gaming takes precedence over other interests and daily activities," and carrying on or escalating "despite the occurrence of negative consequences."

Notice what's doing the work there. Not hours. Not whether you enjoy it. It's control, displacement and consequences, which is roughly the same list you'd use for any habit that's got away from someone.

And there's a threshold most people skip past. WHO says the pattern must cause "significant impairment to a person's functioning in personal, family, social, educational, occupational or other important areas," and would normally have been evident for at least 12 months. A rough fortnight of playing too much isn't the diagnosis. It's a rough fortnight.

WHO is also direct about scale, which is the part that rarely gets quoted: gaming disorder "affects only a small proportion of people who engage in digital- or video-gaming activities." Their advice to everyone else isn't to cut down. It's to stay alert to changes in your own functioning. Which is a more useful instruction than a time limit, and it's the same test the rest of this article keeps coming back to. Check how you feel afterwards, not how long you played.

Advertisement

Can a game help after something genuinely frightening happens?

Everything above is about everyday stress: a long day, a tense inbox, a bad week. But there's a separate line of research on something much sharper, the first hours after a real shock like a car crash. It's the one place a specific game has been tested inside a hospital. And the game is Tetris.

The idea comes from memory science. After a frightening event, your brain spends hours consolidating what happened, and the vivid visual fragments are the part that tends to come back later as intrusive images. Tetris is heavy on visuospatial processing. So the theory is that it competes for the same mental resources while that memory is still settling, leaving the images less sticky.

Iyadurai, Holmes and colleagues put that to a real test in the emergency department of the John Radcliffe Hospital in Oxford. They randomised 71 patients who'd just been in a motor vehicle accident, all within six hours of it. One group got a brief reminder of the accident followed by about 20 minutes of Tetris. The other filled in a written activity log for the same length of time. As the trial reported in Molecular Psychiatry (2018), the Tetris group had 62 percent fewer intrusive memories over the following week than the control group, and the ones they did have tailed off faster. Patients rated it very easy and very helpful.

That's a striking result. But read the rest of the abstract before you repeat it. The difference showed up on intrusion symptoms at one week, not on other post-trauma symptoms, and not at one month. The authors called it a proof-of-concept study and said a bigger trial was needed.

The replications since then are mixed, and that matters. A preregistered multilab study led by Wessel and van Schie, published in Collabra: Psychology (2025), had 433 healthy participants watch a distressing film. Tetris cut the intrusions people reported during the task itself. It didn't reduce them over the following week, which was the effect everyone cared about. And when Broughill, Commins and Hyland had 94 people play on their own phones without supervision, reported in the Journal of Traumatic Stress (2025), one session came out 22 percent lower than control and three sessions came out 13 percent higher. Neither difference was statistically significant.

What it does add to everyday advice is a reason the point about absorbing games is more than a figure of speech. A game that genuinely loads your visual attention crowds other things out for a while. That's handy after a rough meeting. It's just not the same thing as treatment, and the research has been honest enough to show where the line is.

Does it matter why you play, or only how much?

Why matters more, and there's much better evidence on this than on the hours question. One particular reason for playing is the one that tracks with problems.

Hsin-Yi Wang and Cecilia Cheng pooled the motivation research in JMIR Mental Health in 2022, covering 49 reports and 86 independent samples totalling 52,254 participants. They sorted why people play into broad categories and checked each against gaming disorder symptoms.

The ranking is clear. Escapism came out strongest at r = 0.40. Achievement followed at 0.32, immersion at 0.22, and social motivation at 0.20. Down at the bottom, playing out of curiosity to discover things managed only 0.07.

So the motive most people would describe as harmless, wanting to get away from things for a bit, is the one that lines up with trouble. Playing to win, to be absorbed, or to be with friends all track more weakly.

Two caveats before you read too much into that. These are correlations with disorder symptoms, not evidence that escaping causes harm, and the direction could easily run the other way, with people who are struggling reaching for games more. And one moderation result is genuinely odd: the escapism link was much stronger in individualistic societies, at r = 0.47, than in collectivistic ones, at 0.25. The authors put that down to differences in prevention programmes, which is a hypothesis rather than a finding.

The more useful study is newer, and it reframes the whole question.

Agnieszka and Paweł Strojny ran two studies with Polish gamers, published in Frontiers in Psychiatry in 2026, with 259 participants averaging 40 years old and 193 averaging 19. Their argument is that escapism is morally neutral. What separates the good version from the bad one is what they call post-gaming return, meaning whether you can put the game down and step back into your actual life.

Measured that way, the numbers are striking. Post-gaming return correlated with gaming disorder symptoms at r = -0.71 in the first study and -0.58 in the second. It correlated positively with mood after playing at 0.38, with in-game need satisfaction at 0.41, and it predicted higher vitality in both samples.

And here's the bit that settles the argument. When they added post-gaming return to a model predicting disorder symptoms from escapist motivation alone, the variance explained jumped from 0.18 to 0.51 in the first study, and from 0.21 to 0.36 in the second. Knowing how well someone comes back tells you far more than knowing they play to escape.

Same caveats apply. Both studies are cross-sectional, both are Polish samples, and the two age groups are twenty years apart, so treat the exact figures as provisional in the way this page keeps insisting on.

But the practical implication is genuinely useful, and it's different from the advice you usually get.

The question to ask yourself isn't whether you're escaping. Almost everyone is, at least partly, and that's fine. The question is what the twenty minutes after you stop look like.

• Check the landing, not the session. If you close the game and feel roughly ready to deal with your evening, the escape did its job. If you come back flat, irritable, or reluctant to face anything, that's the signal worth taking seriously.
• Notice the difference between playing to feel something and playing to feel nothing. Absorption and numbness look similar from outside and feel very different afterwards.
• Don't moralise the escape itself. Treating any use of a game as avoidance is both unsupported and unhelpful. The research points at the return, not the departure.
• Watch for the drift from one to the other. The same game can be recovery in a normal week and avoidance in a bad one. It's the week that changed, not the game.

This also sharpens the earlier point about quantity. Hours on their own are a poor signal, which is why the Oxford work found so little from playtime alone. How you re-enter your life afterwards looks like the better thing to pay attention to.

Can the game itself be the thing stressing you out?

Yes, and usually it isn't the gameplay doing it. It's the scaffolding around the gameplay: the streak counter, the daily reset, the energy bar, the thing you have to log in for. Those exist to make you return on a schedule, and a schedule is the opposite of the thing that made games restful in the first place.

The section above asked whether using games to cope backfires. This is the other failure mode, and it has nothing to do with your habits. It's built into the product.

Why obligation cancels the benefit

The clearest explanation comes from the foundational work on why games feel good at all. Ryan, Rigby and Przybylski ran four studies published in Motivation and Emotion (2006), applying self-determination theory to game play. Studies one to three had people play and complete measures before and after. Study four surveyed an online community of multiplayer gamers.

Their finding: perceived in-game autonomy and competence were associated with enjoyment and with changes in well-being from before play to after it. In the multiplayer study, autonomy, competence and relatedness each independently predicted enjoyment and future play.

Autonomy is the one that matters here. It means the sense that you're playing because you chose to. A 400-day streak quietly converts that into something else. You're not choosing to play tonight, you're avoiding losing a number. Same game, same mechanics, and the psychological ingredient the well-being effect depends on has been removed.

You can usually feel the switch when it happens. Opening the app stops feeling like a break and starts feeling like clearing an inbox.

What the spending research shows

Monetisation is the sharper end of this, and there's a large dataset on it.

Zendle and Cairns surveyed gamers about loot boxes and published in PLOS ONE (2018), analysing 7,422 respondents after cleaning. Spending on loot boxes was linked to problem gambling severity at an effect size of η² = 0.054. Spending on ordinary in-game items came in at η² = 0.004, more than thirteen times smaller. So it wasn't spending money that tracked with harm. It was specifically the randomised, gambling-shaped kind.

The spending gap across groups is stark too. Non-problem gamblers reported roughly $1 to $5 a month on loot boxes. Problem gamblers reported $20 to $40.

One caveat the authors themselves insist on, and it matters. This is correlational. They say plainly it's unclear whether loot boxes act as a gateway to problem gambling or whether people already prone to it are drawn to loot boxes. Either reading supports the same practical advice, but neither supports saying loot boxes cause anything.

How to tell a restful game from an extractive one

Before you install something as a wind-down habit, check a few things. None of these require you to be an expert:

• Does it punish you for not playing? Streaks, decaying progress and expiring rewards are all the same mechanic wearing different hats. A game that's happy to sit untouched for a fortnight is a better wind-down candidate.
• Does it decide when you play? Energy meters and daily resets hand your schedule to the game. That's autonomy going out the door.
• Can you stop mid-session without losing anything? If quitting costs you progress, you can't actually use it for ten minutes, which is the length most people have.
• Is there anything randomised you can pay for? That's the specific feature the research above singles out.
• Does it notify you? A push notification is the game deciding you should feel something now. Turn them off and see whether you still want to play.

This is also the honest case for plain browser games, and it's structural rather than a sales pitch. Something with no account, no login streak and nothing to buy can't build an obligation loop around you, because there's no hook to hang one on. You open it, play, close it. That's the whole relationship.

And if a game you used to enjoy now feels like an item on a list, the useful move isn't willpower. Break the streak deliberately, once, and notice how you feel a day later. Most people find the answer is relief, which tells you what the streak was actually doing.

Does playing with other people work better than playing alone?

Sometimes. And it depends far more on who you're playing with than on whether the game is multiplayer at all.

Everything above this point has quietly assumed solo play, which is a real gap, because social support is one of the best-evidenced stress buffers there is. If a game puts you in contact with people you like, that's a second mechanism running alongside the attention one. The question is whether it reliably does.

The Oxford team behind the telemetry research quoted earlier came back to this. Ballou, Hakman, Vuorre, Magnusson and Przybylski published a narrative review of 13 proposed mechanisms in Technology, Mind and Behavior (2025), an American Psychological Association journal. Social health is one of the mechanisms they examine, and their summary of it splits cleanly in two.

Multiplayer gaming with friends can strengthen bonds. Playing online with strangers can expose players to toxic environments, suppress social ties, and lead to feelings of loneliness. Same feature, opposite results. They make the same point about needs more broadly: games can satisfy autonomy, belonging and mastery, or trigger coercion, failure and loneliness.

There is one large trial worth adding, because it takes the social idea and actually tests a structured version of it. Kenji Yokotani, Masanori Takano, Nobuhito Abe and Takahiro Kato ran a randomized controlled trial with 1,105 players of a massively multiplayer online game, published in JMIR Serious Games in 2025. The 548 people in the experimental group were given small social tasks inside the game, things like ringing the doorbell on other players' virtual rooms on weekdays, with the 557 in the waitlist group carrying on as normal.

It worked, and barely. Depression scores in the experimental group fell from 8.78 to 7.45 over three months, against 8.1 to 7.24 in the waitlist group. The standardised mean difference was 0.12, which is a small effect by any conventional reading.

Take that as calibration rather than discouragement. This is a properly designed trial with over a thousand people and a deliberate social intervention, and it still landed on a small effect. Anyone promising you that multiplayer gaming will transform your mental health is selling something the best available evidence does not support.

So "play with other people" isn't advice yet. It's a fork.

The sharper finding comes from a study that measured the fork directly. Mandryk, Frommel, Armstrong and Johnson surveyed 300 World of Warcraft players for Frontiers in Psychology (2020), looking at whether in-game relationships actually translated into feeling less alone.

Both kinds of player built social capital. Players with harmonious passion, meaning they played because they wanted to and could stop, showed strong ties (bonding social capital β = 0.604, bridging β = 0.798). Players with obsessive passion, meaning play felt compulsory, built ties too, just weaker ones (β = 0.140 and 0.157).

Then the two groups came apart completely:

• Harmonious passion went with less loneliness (β = -0.307) and predicted wellbeing (β = 0.315).
• Obsessive passion went with more loneliness (β = +0.270) and had no significant relationship with wellbeing at all (β = 0.007, p = 0.912).

Read that carefully, because it's the least obvious thing on this page. Having gaming friends is not the same as feeling less alone. Both groups had the relationships. Whether those relationships did anything depended on why the person was there. The authors traced the route explicitly: harmonious passion improved wellbeing by building close ties which then reduced loneliness. For obsessive players that chain simply didn't run.

Which is the same harmonious-versus-obsessive split from the recovery section above, showing up again on the social side. It's turning out to be the more useful variable in this whole topic than any genre label.

The usual caveats, and this article keeps insisting on them. It's cross-sectional and self-reported, so it can't tell you which way the causation runs. And it's 300 players of one game, 261 of them men, average age 30.7, with 70 percent playing daily. That's a committed population, not gamers generally. Take the direction, not the size.

What that leaves you with, practically:

• Play with people you actually know if you want the social part to do any work. That's the condition under which the evidence points positive.
• Be careful with anonymous competitive lobbies as a stress remedy specifically. They might be fun. But toxicity and loneliness are the documented downside risk, and you're supposed to be unwinding.
• Notice whether you're playing because you want to or because you feel you have to. That single question predicted more here than hours, genre or whether other people were present.

If none of that appeals and you'd rather decompress alone, that's a perfectly good answer. Solo absorption is what most of this article is about, and the flow state guide covers why it works.

Does your body agree with what you report?

Not always. And the gap between the two is the most interesting thing in this whole area.

Nearly everything above rests on self-report. Somebody plays a game, then fills in a questionnaire about how stressed they feel. That's useful, but it's also the weakest link in the chain, because people are unreliable narrators of their own arousal and they can usually guess what a study is hoping to find. The harder test is whether the body cooperates.

A 2025 lab experiment ran exactly that test. Gary Wagener and colleagues at the University of Luxembourg put 82 participants through the Socially Evaluated Cold Pressor Test, a standard way of reliably stressing someone in a lab, then had them play about 25 minutes of A Plague Tale: Requiem on a PlayStation 5. Half played violent passages, half played non-violent ones from the same game. The work was published in the International Journal of Psychophysiology.

Cortisol and heart rate dropped in both groups. Heart rate variability went up, which is the direction you want, since it generally tracks with the body settling rather than bracing.

Here's the part worth sitting with. The people playing the violent passages reported feeling more stressed and more aggressive than the non-violent group, while their physiology was going the other way. Same session, two measurements, opposite answers.

Three things follow from that, and they cut in different directions.

• The calming effect is not just people being agreeable in questionnaires. Something measurable happens. That's a genuine strengthening of the case this page has otherwise been careful about.
• Feeling wound up doesn't necessarily mean you are. If you finish a tense game session with your heart still going and conclude it made things worse, you might be reading the wrong instrument. Or you might not be. One study on one game is not a rule.
• Self-report and physiology measure different things, so pick which one you care about. If your goal is sleeping better, the body is the thing that matters. If your goal is feeling less awful this evening, your own read counts for more.

The caveats are the usual ones and they are not small. It's 82 people, one game, one lab, one acute stressor that looks nothing like a bad month at work. Cortisol also falls on its own after a stress spike, so some of that drop is just time passing rather than anything the game did. The design compares violent against non-violent play rather than against sitting quietly, which limits how far you can push the conclusion.

Still, it's the sort of evidence this topic mostly lacks. Physiological measures are expensive and awkward to collect, which is why so much of the literature leans on questionnaires. When someone does go to the trouble, and the body disagrees with the questionnaire, that's worth knowing before you trust any percentage on this subject.

Does the sound the game is making change anything?

It can, and it's the one variable on this page you can change in a single click. Most people never touch it.

The section above landed on cortisol as the harder test, the measure that doesn't care what you write on a questionnaire. So here's the same instrument pointed at something smaller than the whole game.

Hébert, Béland, Dionne-Fournelle, Crête and Lupien ran the experiment and published it in Life Sciences (2005). Two groups played the same video game for ten minutes. One group had the game's built-in music on. The other played in silence. Everyone gave four saliva samples across the session so the researchers could track cortisol rather than ask about feelings.

The group with the music on showed higher cortisol. Same game, same ten minutes, same task. The soundtrack was doing measurable work on their stress hormones, in the wrong direction.

Now the caveats, because this one gets stretched further than it should. It's a 2005 study on one game with pop and techno built-in audio, which is a specific kind of music rather than game music in general. A slow ambient soundtrack is not techno, and nothing here says it behaves the same way. And the design compares audio against silence, not against a calmer alternative track.

So the honest read is narrow. Loud, driving, fast game audio can raise your cortisol even while you feel like you're relaxing. That's it. It isn't an argument for playing everything on mute.

But it's a useful narrow finding, because browser games are close to the worst case. The audio tends to be short loops built to grab attention rather than hold a mood, the volume is often mastered louder than the rest of your machine, and on ad-supported pages it sits next to whatever an ad break decides to play at you. None of that was designed to settle your nervous system.

Three things worth trying, none of which cost you anything.

• Run one session muted and compare. Not as an experiment with a result, just as a thing you notice. Plenty of people find a puzzle game reads completely differently without the loop underneath it.
• Substitute rather than subtract. Mute the game and put on something you'd choose deliberately. You keep the absorption the game gives you and drop the audio you didn't pick.
• Treat a sudden volume jump as a signal. If a game or the page around it gets loud without you doing anything, that's the attention-grabbing design the earlier section warned about, now operating on a channel you weren't watching.

This connects back to the point about extractive design. A game that fights for your attention through sound is using the same playbook as one that fights for it through streaks and notifications. The difference is that the sound is easier to switch off, and switching it off costs you nothing you actually wanted.

What are the best games for unwinding?

For a calm session, reach for something with no harsh timer and no opponent, just steady, satisfying progress:

• Solitaire, the classic no-pressure card game.
• Mahjong solitaire, soothing tile-matching at your own pace.
• A gentle nonogram, which rewards calm, steady logic.
• A relaxed game of snake or a sliding puzzle when you want something light.

The common thread is progress without pressure. Play for the calm, not the score. For more low-key options, our roundup of puzzle games for adults has plenty to try.

How long should you play to de-stress?

Keep sessions short and intentional. A ten to twenty minute break to reset is restorative, while hours of play, especially late at night, can push your bedtime back and add stress rather than remove it. The sleep evidence for that is narrower than it's usually presented, which the next section gets into. Set a rough time limit before you start so a quick reset does not turn into a long detour.

Then pay attention to how you feel afterward. If a session leaves you calmer, it worked. If it leaves you wired or frustrated, switch games or stop. Gaming for stress relief is about the feeling, not the finish, and a short, well-chosen break is the whole point.

Does a quick game work as a break during the workday?

Better than doing nothing, and that's the comparison almost nobody actually runs.

Michael Rupp and colleagues at the University of Central Florida published "Searching for Affective and Cognitive Restoration: Examining the Restorative Effects of Casual Video Game Play" in Human Factors in 2017, volume 59, issue 7, pages 1096 to 1107. It's worth more than most studies on this page because it put three things against each other inside one experiment instead of testing a game on its own. Participants ran a draining vigilance task, then got randomly assigned to a casual video game, a guided relaxation activity, or a plain passive break.

The game won on mood. Players showed greater engagement and greater affective restoration than the relaxation condition. And the plain break, the one where you just stop and sit, slightly decreased affect.

Read that last part twice, because it's the genuinely counterintuitive bit. Doing nothing didn't leave people flat. It left them slightly worse off than when they started. If you've ever stared at a wall for five minutes hoping to feel refreshed and come back grumpier than you left, that's the effect showing up in a lab.

It also quietly explains why "just take a break" is such useless advice. A break isn't one thing. An empty one and an absorbing one do different things to you, and the study found they point in opposite directions.

Will a game break make you work better afterwards?

Probably not. This is where the honest answer parts company with the productivity advice, so it's worth being clear about.

Rupp's own conclusion is blunt. Playing a casual game briefly can restore your affective abilities, and then the paper ends by saying future research is needed to find activities capable of cognitive restoration. None of the three conditions fixed cognition. Not the game, not the relaxation exercise, not the break.

The broader literature lands in the same place. Paul Albulescu and colleagues at the West University of Timisoara published "Give me a break!" in PLOS ONE in 2022, a meta-analysis pooling 22 independent samples and 2,335 participants. Micro-breaks produced small but statistically significant gains in vigor (d = .36) and reductions in fatigue (d = .35). The effect on overall performance was d = .16, and it wasn't statistically significant at all (p = .116).

Two more details sharpen it. Performance effects turned up only for tasks with lower cognitive demand. And a meta-regression found that the longer the break, the bigger the performance gain, with the authors concluding that recovering from highly depleting tasks may need more than ten-minute breaks.

So a five minute game between two hard problems will probably make you feel better and won't make you think better. Those are two different products, and most advice on this sells you the first while promising the second.

What this means for your actual workday

• Use a game break for mood, not for output. If you're stuck on something hard, a quick round won't unstick you. If you're irritable and about to be unpleasant in a meeting, it might genuinely help. Pick the right tool for the right problem.
• Don't sit and stare instead. The passive break was the only condition that actually decreased affect. Whatever you do with the ten minutes, doing nothing with them is not the safe default it sounds like.
• Scale the break to the drain. The meta-analysis is explicit that heavy cognitive depletion probably needs longer than ten minutes. A micro-break is sized for a normal afternoon, not for the day you've been in back-to-back problem solving since morning.
• Match the game to the gap. Something with no timer and no opponent fits a work break far better than anything competitive, for the reasons the earlier sections cover. A round that makes you tense has cost you the break.
• Watch the clock deliberately. The whole benefit here is measured in minutes. A break that eats an hour isn't the thing these studies tested, and it isn't the thing they found works.

And notice how well this fits the rest of the page. The pattern keeps repeating: the effect on how you feel is real and reasonably well evidenced, while the effect on what you can do is smaller, patchier, and usually oversold by whoever is selling something. If it's the thinking side you're actually trying to fix, our guide to games to improve focus is a better starting point than this one, and it's honest about the same gap.

Is a game better than a meditation app?

Somebody ran that exact comparison, which is rarer than you'd expect. The short answer is that the game did better on one measure, tied on another, and pulled ahead over a week. The longer answer is more useful, and it comes with the usual caveat about sample size.

Emily Collins, Anna Cox, Caroline Wilcock and Geraint Sethu-Jones published "Digital Games and Mindfulness Apps: Comparison of Effects on Post Work Recovery" in JMIR Mental Health in 2019. They ran it twice: a lab study with 45 people, then a five-day field study with 20 working professionals. Both compared a puzzle game, Block! Hexa Puzzle, against the Headspace mindfulness app.

In the lab, participants did a demanding work task first, then got one of the conditions. The game was the only one that significantly raised energetic arousal afterwards, meaning people felt more awake and less drained. Headspace didn't move it. On the separate measure of recovery experience, though, nothing separated the conditions at all.

So on that first pass it's a split decision. Both left people feeling equally recovered. Only the game left them with energy.

The field study is where it gets interesting, because it ran across a working week rather than a single afternoon. Recovery experience didn't stay flat. It moved in opposite directions depending on the condition: game scores climbed as the week went on, while mindfulness app scores fell.

Worth being careful about what that does and doesn't show. Twenty people over five days is small, the paper is now several years old, and nobody has replicated it at scale. It's a signal, not a verdict, and this page has been fussy about that distinction elsewhere so it applies here too.

But the direction fits everything above. One more finding from the lab study ties it together: how much people enjoyed the activity correlated strongly with how recovered they felt, at r = .69. Enjoyment did not predict the energy change, only the sense of recovery. Which is the Oxford result again, arriving from a different angle. What you get out of a break tracks with whether the thing suited you, not with the category it belongs to.

That reframes the question you should be asking. It isn't games versus meditation as a general contest, because the evidence doesn't support declaring a winner and neither do the authors.

• If you finish work drained and still need to function, the game has the edge on paper. It's the one that put energy back rather than just calm.
• If a meditation app has never stuck for you, that's informative rather than a personal failing. An intervention you don't enjoy scores worse on the measure that matters most, and the declining scores across that week are consistent with people quietly disengaging from something they were tolerating.
• If meditation does work for you, keep doing it. Nothing here says otherwise. The two conditions tied on recovery experience in the lab.
• Don't run both as a chore. Stacking two recovery activities you don't like is a good way to spend your evening on admin.

The honest summary is that the best recovery activity is a real one you'll actually return to. The research keeps landing there no matter which end it starts from.

What about games that use your breathing as the controller?

This is the one corner of the topic with better evidence than anything else on this page. It's also the one where the obvious conclusion is wrong, because the thing that makes these games work is in tension with the thing that makes them games.

Start with the underlying technique rather than the games. Heart rate variability biofeedback means breathing slowly, usually around six breaths a minute, while a live readout shows you what your heart is doing in response. You're not guessing whether it's working. You can see it.

V. C. Goessl, J. E. Curtiss and S. G. Hofmann pooled the evidence in "The effect of heart rate variability biofeedback training on stress and anxiety: a meta-analysis", published in Psychological Medicine in 2017 (vol. 47, issue 15, pp. 2578 to 2586). They found 24 studies covering 484 participants, and the effect was large: Hedges' g of 0.81 comparing before against after, and 0.83 comparing biofeedback against a control condition. Efficacy held steady across study year, risk of bias, gender split, number of sessions, and whether participants had a diagnosed anxiety disorder.

That's a bigger and more consistent effect than almost anything else cited on this page. Now the caveat this page owes you, because it's the same one applied everywhere else here.

484 participants across 24 studies is an average of about 20 people per study. Large effects from small samples are exactly the pattern that shrinks when somebody runs it properly at scale, and it's the reason the percentages section above exists. Treat the direction as well supported and the size as optimistic.

So far none of that involves a game. Here's where it gets interesting.

Venkata Nitin Chakravarthy Gummidela, Dennis R. da Cunha Silva and Ricardo Gutierrez-Osuna ran "Evaluating the Role of Breathing Guidance on Game-Based Interventions for Relaxation Training" in Frontiers in Digital Health in 2021. Two experiments, 30 participants then 36 in an ambulatory setting, crossing two variables: whether the breathing guidance came from live biofeedback or a fixed pacer, and whether it was wrapped in a game or not.

The game did two things well and one thing badly, and all three matter.

• People enjoyed it more, by a wide margin. F(1,20) = 24.49, p < 0.001, with an eta squared of 0.55. That's not a marginal preference.
• People actually used it more in daily life. In the ambulatory study, where participants chose when to engage on their own time, the game conditions got picked more often (p = 0.027). Adherence is usually where relaxation training dies, so this is the whole ballgame for most people.
• But people breathed worse while playing it. The non-game conditions hit effective breathing in roughly 85 percent of sessions. The game conditions managed about 50 percent.

Read those three together and the trade-off is blunt. Wrapping the breathing in a game roughly doubles the chance you'll come back to it and roughly halves the chance you're doing it right while you're there. The game is competing with the exercise for the same attention, which is the mechanism, not a flaw in the study.

The authors didn't leave it there, and the follow-up detail is the practical one. Only the game-plus-biofeedback condition produced a significant drop in breathing rate from before to after, t(5) = 5.94, p = 0.001. And training mattered a lot: participants who'd been trained properly hit the target 80 percent of the time against 53 percent for those who hadn't. The skill is learnable. The game just isn't a good place to learn it.

Which gives you a sequence rather than a choice:

• Learn the breathing without the game first. A plain pacer, a few sessions, until slow breathing stops taking concentration. This is the boring part and it's the part that carries the effect.
• Then use the game to keep the habit alive. Once the pattern is automatic, the thing competing for your attention costs you much less, and you get the adherence benefit that the plain version never delivers.
• Don't expect a breathing game to teach you from scratch. Half the sessions missing the target is a lot of time spent feeling productive while getting the dose wrong.
• Watch out for the readout becoming the score. Biofeedback works because you stop striving. A game that turns your heart rate into a leaderboard is pointed the wrong way, and the section above on extractive design applies here with extra force.

Two honest limits. Both experiments are small, 30 and 36 people, which is the same objection this page raised against the meta-analysis and it doesn't get suspended because the result is inconvenient. And this research is about relaxation training specifically, not about whether a puzzle game unwinds you after work. Those are different questions with different evidence, and the rest of this page covers the second one.

Worth noting where this sits relative to everything else here. Most of this page is about games that reduce stress as a side effect of being absorbing, which is the flow state mechanism. Breathing games work through a completely different route: they're a delivery system for a physiological technique that already works on its own. That's why the evidence is stronger, and also why the game part turns out to be the weakest link in the chain.

Does playing before bed actually hurt your sleep?

Less than you've been told, and the real risk isn't the one people warn you about.

"Don't game before bed" is repeated everywhere, including twice further up this page, so it's worth checking whether the evidence matches the confidence. Somebody measured it in a sleep lab. Researchers at Flinders University in Adelaide had adolescents play Call of Duty 4 before bed on one night and watch a nature documentary on another, then compared how long they took to fall asleep. The results appeared in the Journal of Clinical Sleep Medicine and were summarised by the American Academy of Sleep Medicine.

After the shooter, they took a median of 7.5 minutes to fall asleep. After the documentary, 3 minutes. So gaming did delay sleep onset, and the direction is real.

But look at the size of it. Four and a half minutes, after a violent action game, which is about as arousing as presleep gaming gets. The researchers said so themselves: the increase was small and may have few perceptible ramifications for sleep. If a calm puzzle costs you a fraction of that, it isn't the thing standing between you and a good night.

Now the finding that actually matters, because it moves the problem somewhere else entirely. Players also reported feeling less sleepy afterwards. The researchers flagged that as the real concern, since feeling wide awake at midnight is exactly what talks you into one more round.

Keep the sample honest, though. That was 13 male teenagers who were good sleepers, playing a 50 minute session in a lab. Small, narrow, and not obviously about a 40-year-old winding down with mahjong. The authors said as much. Treat it as a reason to stop assuming presleep gaming is uniquely harmful, not as clearance to play until 2am.

Physical health and mental sharpness go together, so check your numbers with the free tools at HealthCalcAsia.