Opinion: So what *does* the science say about games and violence?

In the wake of renewed debate about violence and video games, Gamasutra contributor Katherine Cross dons her academic cap and examines what the sciences have to say on the subject.

In the wake of renewed debate about violence and video games, I suppose I should put that sociology degree to work and explain, roughly, the academic state of play. It’s worth noting that there’s next to nothing that supports what I like to call the “monkey-see-monkey-do” model of media influence, i.e. a piece of media shows a violent act, and viewers/players go out and commit the same act.

The heart of the debate in the social sciences is, instead, about “aggression.” As the American Psychological Association, put things: “all existing quantitative reviews of the violent video game literature have found a direct association between violent video game use and aggressive outcomes.” But what, exactly, are “aggressive outcomes”?

As I often tell students, the simplest words are the most devilish in science (consider how a sociologist might have to confront the idea of “love” and you’ll see what I mean). Such terms have to be rigorously defined within a single study, but that careful definition may differ from how other studies may concretise the term. "Aggression" is one such word.

This is why, in their 2015 “Resolution on Violent Video Games” the American Psychological Association had to devote a great deal of time to dealing with just what that meant--notably, the same resolution that made the conclusion about “aggressive outcomes” I cited above.

“The violent video game literature uses a variety of terms and definitions in considering aggression and aggressive outcomes, sometimes using "violence" and "aggression" interchangeably, or using "aggression" to represent the full range of aggressive outcomes studied, including multiple types and severity levels of associated behavior, cognitions, emotions, and neural processes. This breadth of coverage but lack of precision in terminology has contributed to some debate about the effects of violent video game use.”

The statement goes on to note that problems arise due to the interdisciplinary nature of the field (criminologists and epidemiologists may mean different things when they say “aggression” resulted from x stimulus). But this all culminates in the following conclusion:

Thus, all violence, including lethal violence, is aggression, but not all aggression is violence. This distinction is important for understanding this research literature, which has not focused on lethal violence as an outcome. Insufficient research has examined whether violent video game use causes lethal violence.

Emphasis emphatically mine. This is why the APA’s Society for Media Psychology and Technology issued a separate statement about the matter in 2017, underlining that point at some length. “Aggression is a much broader category of behaviors, including many very mild behaviors and some (such as sports aggression, competition, or debating) that are socially sanctioned,” they note, adding that it’s important for the press to cast a wide net when soliciting expert opinion on the subject. They note that confirmation bias is often at play in media reporting, looking for evidence of videogame use when young men commit mass shootings, but not when older men or the vanishingly rare female mass shooter is involved (and, one might add, when non-white shooters are culpable).

Even more importantly, the Society added some details about what “aggressive outcomes” are actually measuring:

“A wide body of research has examined the impact of violent video games on relatively minor acts of aggression, such as the administration of unwanted hot sauce to make food too spicy, making someone put his or her hand in freezing ice water or bursts of white noise in laboratory experiments. These studies have resulted in mixed outcomes, some reporting evidence for significant effects, and others do not.”

Everything is data, and all data is valuable. I won’t cast aspersions on this research by saying it measured nothing of significance. What I can say is that the connection between overt violence and video gaming is poorly established and, at best, requires more study. It may be that playing a violent video game may make you nastier for a few minutes, but it’s hard to say that it’ll turn a perfectly ordinary person into a killing machine. The hurdles involved are so numerous and so high, it’s hard to fathom the nudge-effects of media getting a disinclined person over them.

Even flashy new studies that establish a link to “aggression” are not beyond severe criticism. One study from 2014 found that, over the course of three years, children who played violent videogames became more “aggressive” in social situations--interpreting ambiguous acts, such as accidental shoves in a hallway, as hostile. There was also actual violence recorded: hitting and shoving, for instance. But, as other scholars pointed out, the study relied on the children’s own reports of whether a videogame was violent or not; it also lacked a critical statistic known as “effect size” which would account for how much a given variable (in this case, violent videogame play) influenced another (in this case, aggression).

And this was one study. To say the field is, at best, muddled is an understatement. It’s more accurate to say that there’s an emerging consensus that videogaming’s impact on people’s aggression is highly granular and not, in any way, directly tied to overt criminality--much less extreme antisocial behaviour like terrorism or mass shootings. Such extreme acts are always multicausal. There’s a conversation to be had about how our media promotes violence as a solution to every problem (a particularly manly solution at that). But that discussion has to account, from the start, for the fact that many other countries absorb the same media without the crisis of mass shootings that so afflict America.

Influencing a belief is not the same as causing an action.

There are some who say that this proves any discussion about media impact is flawed, which informs some of the brutal defensiveness that greets criticism of, say, racism in videogames. But that’s equally wrongheaded. Media is persuasive; try telling advertisers that the billions they spend on commercials and spreads are wasted. Media may not be hypnotic, but it can still influence how people see the world (which is different from what they may directly do to the world). Having one's rape-myth acceptance heightened is different from one going out and committing a rape; but the former is still worth studying and evaluating, especially as findings are mixed.


Talking about socialization to a wide audience can be a frustrating exercise for a social scientist, as it’s very easy to misunderstand. Those misunderstandings infect even the loftiest discussions about media influence. People can act as if socialisation theory is absolute: you are taught x by an authority or mass media, therefore you believe x for the rest of your life. Any example that deviates from this absurdly strict concept of socialisation is seen as proof that we’re all perfect individuals, free from societal influence. But the picture is more complex.

A person is not a passive vessel into which ideas are poured. We are all, always, engaged with what we are learning. Influenced by it, yes, but also remixing and interpreting it for our own purposes, in ways that often exceed or confound the intentions of a work’s creators or an instructing authority figure like a parent, cleric, or teacher. But socialisation still sets parameters; fields of discourse from which few people deviate wildly.

Socialization is a dialogue, not programming. This is why, for instance, online games foment transgender transitions, how speedruns and mods are born, why Birdo has a huge trans fandom, or Bayonetta has a legion of feminist fans. There are many ways to interpret a work, and it’s difficult to gauge how masses of people will be changed by it.

That’s not to say that social-scientific research can’t address itself to that question, but rather that it must do so modestly, accounting for that give-and-take. For the most part, it does. The problem comes from overheated interpretations of that necessarily modest work.

This is the challenging terrain on which discussions about media influence are often had, where we discuss the importance of creating new female/non-white role models, or of queer representation in our games. It’s with the understanding that we may not know the exact impact of it all for years to come, while also valuing those players who suddenly feel included in the here and now and seeing that as valuable for its own sake. It’s not videogaming per se, but to look at the reaction to Black Panther is to see something beautiful resulting from inclusive art.

There is power, yes, in creating a work for people to interpret; but there’s a parallel, capillary power in what those people do with it. It’s something that will always contain at least a few surprises.

That last bit is what truly fascinates me.

Katherine Cross is a Ph.D student in sociology who researches anti-social behavior online, and a gaming critic whose work has appeared in numerous publications.

Latest Jobs

IO Interactive

Hybrid (Malmö, Sweden)
Gameplay Director (Project Fantasy)

Arizona State University

Los Angeles, CA, USA
Assistant Professor of XR Technologies

IO Interactive

Hybrid (Copenhagen, Denmark)
Animation Tech Programmer

Purdue University

West Lafayette, IN, USA
Assistant Professor in Game Design and Development
More Jobs   


Explore the
Advertise with
Follow us

Game Developer Job Board

Game Developer


Explore the

Game Developer Job Board

Browse open positions across the game industry or recruit new talent for your studio

Advertise with

Game Developer

Engage game professionals and drive sales using an array of Game Developer media solutions to meet your objectives.

Learn More
Follow us


Follow us @gamedevdotcom to stay up-to-date with the latest news & insider information about events & more