Hikikomori: Why At-Risk Recluses Gravitate to Story and Strategy Games

Ask a few hundred Japanese gamers to name the games they love, and the list runs where you would expect: role-playing epics, sprawling adventures, the odd farm to tend at your own pace. Now split that same crowd by how much each person wants to withdraw from the world around them. The favourite games barely move. The genres the low-withdrawal group reaches for look, on the raw numbers, a lot like the genres the high-withdrawal group reaches for. Which is, in its way, the first surprise in a study that set out expecting a difference.

The work, published in PLOS Mental Health, was after something practical: how you might build a game that helps people who have pulled back hard from social life, rather than one that pushes them further in. To get there, the team first had to know what these players actually enjoy.

The condition they are circling is hikikomori, a severe and prolonged social withdrawal first named in Japan four decades ago and now documented from Italy to Brazil to the United States. It is worth being careful with the word, because the study is. The researchers, led by a group at the Nara Institute of Science and Technology, did not recruit people with a clinical diagnosis. They measured risk, using a questionnaire built around the idea of the NEET (not in employment, education or training) that treats withdrawal as a spectrum of leanings, uncertainty about the future, a pull toward not working, a sense of not being needed, rather than a yes-or-no state. More than 500 adults filled it in, and a 104-point cut-off split them into higher-risk and lower-risk groups, roughly half in each.

So: two groups, one questionnaire, and a simple question. Does where you sit on that scale predict the kind of game you love?

On the genres, statistically, no. The team pulled favourite-game names from participants, mapped them to genres through a game database, and ran the comparisons. After correcting for the many tests involved, nothing significant turned up, not for genres, not for the finer-grained gameplay keywords either. Role-playing, action and adventure games topped both lists. If you were designing purely from this table, you might conclude that building for a withdrawn player looks much like building for anyone else.

The counts do tilt in places, and it is tempting to read into them. Action, puzzle and shooter games got mentioned a bit more by the lower-risk group. The authors reach for an explanation borrowed from earlier research: people at higher hikikomori risk, one study found, gave up sooner on a hard task after being told they had failed. Games that punish you repeatedly might, the reasoning goes, wear this group down faster. Might. It’s a plausible thread, not a measured effect, and the paper acknowledges that the genre numbers themselves never reached significance.

Where the Preferences Actually Showed Up

The interesting stuff came not from counting genres but from reading what people wrote. Alongside the tick-box data, participants answered open questions about what they liked and disliked in single-player, multiplayer and online games, and three coders worked through those answers looking for themes. Here the two groups did start to part ways, though in reported feeling rather than in any tidy statistic. Higher-risk players wrote warmly about playing at their own pace, about worlds they could sink into, about characters and stories worth following. They spoke positively of action and strategy when the challenge felt fair. And when the talk turned to other people, a clear lean emerged: cooperation good, competition much less so. Several described a real fear of being a nuisance to teammates, of what an opponent might do after a loss. One theme the researchers kept meeting was the sting of toxic behaviour in online play, and a sense of helplessness about stopping it.

None of this is a stereotype of the antisocial recluse, and that may be the point. The withdrawn players were not uninterested in other people. They seemed to want connection on terms they could manage, a guild to belong to, a character to grow fond of, a friend to share a walk-around game with, minus the exposure that competitive play brings. Story-driven and strategic games, played cooperatively or alone, fit that shape.

What the Title Claims and the Numbers Do Not

All of which sits a little against the study’s own title, and against the press release announcing it, both of which describe these preferences as associated with a higher risk of hikikomori. Read quickly, that sounds like a measured link: like the games and the risk rise together. The tested comparison found no such thing. The association is a qualitative reading of what higher-risk players said, not a number that separates them from everyone else, and the paper is scrupulous about the distinction even where the headline blurs it.

There is a second caveat the authors flag themselves, and it matters more than it looks. This study asked what withdrawn players enjoy. It did not ask whether that enjoyment is good for them. Internet gaming disorder, hours logged, the line between a healthy habit and a compulsive one, none of it was measured. A design that keeps a lonely person happily glued to a screen is not obviously a design that helps them back into the world, and the paper is careful not to claim otherwise.

Then there is the sample itself. The team recruited through a crowdsourcing platform, on the reasoning that people avoidant of conventional jobs would gravitate to freelance gig work, which is a shrewd way to reach a hard-to-find group. It also means the recruiting method leans on the very tendencies the risk scale measures, and the whole thing is Japanese, adult, and self-selected. Preferences may not travel. The researchers say as much.

What survives all that is still worth having. If you wanted to build a game as a first, gentle step toward reconnection, aimed at people for whom other humans feel like a hazard, this points somewhere: short sessions you can pick up and drop, a world and a story to inhabit, progression that rewards without punishing, cooperation offered rather than competition forced. Maybe even something that nudges you out the front door, a few high-risk players mentioned enjoying location-based games like Pokemon GO and Dragon Quest Walk, the sort that only work if you go outside. Whether any of that actually helps is the next study, not this one.

Frequently Asked Questions

Does liking cooperative or story-driven games make someone more likely to become a recluse?

Liking cooperative or story-driven games does not make someone more likely to become a recluse, and this study cannot show that it does. It is a one-time survey that measures preferences and withdrawal risk at the same moment, so it captures how they sit together, not whether one leads to the other. Its statistical test actually found no significant link between social-withdrawal risk and game genre at all.

What did the study actually find, if the genre numbers showed no difference?

What the study found was that the difference showed up in what players wrote, not in which genres they picked. When higher-risk players described their favourite games in their own words, they leaned toward cooperation over competition, warmed to immersive worlds and stories, and were vocal about disliking toxic or competitive online play. The genre counts themselves looked much the same for higher- and lower-risk players.

Why does the gap between the study’s title and its statistics matter?

The gap between the title and the statistics matters because the title says these preferences are associated with higher withdrawal risk, while the study’s own numerical comparison found no significant association. The link it reports comes from a thematic reading of open-ended answers, which describes a group rather than measuring a difference. Reading the headline as a measured effect would overstate what the data support.

Could these findings be used to design games that help socially withdrawn people?

These findings could help design games as a gentle first step toward reconnection, which is exactly what the researchers were aiming at: short sessions, absorbing stories, rewarding rather than punishing progression, and cooperation offered instead of competition forced. But the study only mapped what withdrawn players enjoy, not whether that enjoyment is good for them. Whether such games actually help, or simply keep someone at the screen, is a question for future research.

Who counts as being at risk of hikikomori in this research?

Being at risk of hikikomori here does not mean having a diagnosis. The study surveyed more than 500 Japanese adults and used a questionnaire that scores withdrawal as a spectrum of leanings, such as uncertainty about the future and a sense of not being needed, then split them at a cut-off into higher- and lower-risk groups. So the findings describe risk tendencies among Japanese adult gamers, not a clinical population, and may not carry over to other countries.