If you’ve spent any time playing Dead Island 2, chances are you’ve noticed the game’s progressive damage system. The Fully Locational Evisceration System for Humanoids, or FLESH, as developer Dambuster Studios call it, is a procedural tool that makes dismembering, melting, or burning zombies look more realistic, as signs of trauma correspond to the attacks you perform, visibly chewing through skin, muscle, organs, and bone. Of course, Dead Island 2 applies all this gore to schlocky, slapstick effect. But FLESH may make you wonder how such gruesome detail might translate to games with more serious themes.
Questions around violence in games have a long history, spanning tabloid moral panics to concerted academic research. While the topic of whether playing violent games may lead to aggressive behavior in real life is still hotly debated, studies tend to show that any correlation is at most minuscule. Yet with the progress of visual fidelity in games, from the FLESH system to the recent trailer for Unrecord, which some thought looked too lifelike to be true, it’s no surprise if the question circles around again.
Aaron Drummond, a senior lecturer at the School of Psychological Sciences at the University of Tasmania (and coauthor of the study linked above), believes that while the topic demands additional research, if increasing realism in game violence did lead to more aggressive behavior, the signs should already be present.
“One would expect to see three things,” he explains. “One, an increase in the number of studies showing an effect of violent content on aggression; two, an increase in the effect sizes of violent games on aggressive behavior; and three, an increase in assaults and violent crimes.” None of these things have happened, he adds, with data in fact trending in the opposite direction.
Paul Cairns, head of the Department of Computer Science at the University of York in the UK, has a similar view. “My instinct is that if violent video games really made people violent, we would be going to hell in a handcart right now,” he says. Cairns has explored the concept of “priming,” or the idea that game violence can somehow alter how we respond to violence elsewhere, potentially leading toward violent behavior. There’s no obvious evidence of priming, he says, and “if you manipulate the realism of games, it really doesn’t lead to any change of priming at all.” If there’s any path from playing games to violent behavior, then, it’s not merely down to violent content. “There’s got to be something else going on there.”
Despite past research, though, it’s impossible to know for sure that increased realism won’t have a negative impact, Cairns says, simply because we’ve never seen the current levels of realism in interactive media before. Yet humans—at least adults—are very good at understanding what’s real and what isn’t, he continues, “which is why [some people] can bear a horror film but can’t even watch people have an injection.” So as long as we understand we aren’t taking part in a real scenario, it seems unlikely that even a highly realistic simulation will spark problematic behavior.
Visual fidelity isn’t the only consideration when it comes to the impact of violence, however. The fact that we enact violence in games rather than merely observe it, as in film and TV, makes it a different proposition, as do structures that have us repeat acts of aggression over and over. Indeed, if you enjoyed Dead Island 2’s bloody displays, you might also have found that the novelty wears off after a few hours. Eventually, the sight of blistered skin and broken legs begins to feel mundane. When so many games are built around combat loops of this kind, do we become desensitized to the impact of violence?
It’s certainly possible, Drummond says, potentially resulting in “decreased emotional and physiological reactivity to the violence one witnesses.” Yet that’s not necessarily a problem. “For instance, desensitization is useful if you want to help someone get over a phobia,” he explains, “which clinicians now use VR to do.” Plus, no “sensible player” generalizes in-game violence to real-life contexts without grasping the moral and legal implications.
So viscerally violent games don’t turn us all into murderers, or even people who don’t flinch at the horrors of violence. But can games convey serious themes around violence, instead of merely satiating a player power fantasy? It depends on how the game approaches using violence as a narrative mechanic. For example, although you may initially feel a sense of awe or guilt in trying to kill the creatures in Shadow of the Colossus, Cairns explains, once you fail a few times on a particular challenge, “it becomes a puzzle system. It doesn't matter how realistic it looks.”
Horror games also wrestle with this problem, as no matter how disturbing a scene may be at first sight, repeated exposure deadens its impact. “We’re very capable of running at these different levels of a symbol that is a game versus an aesthetic experience,” Cairns says. “Or even seeing the symbolism through the aesthetic experience.”
Even The Last of Us, often cited as an example of powerful storytelling in games, has to contend with this phenomenon. The game’s recent TV adaptation perhaps only highlights the issue, since acts of violence there are more naturally sporadic and heavy with consequence, in contrast to the game in which Joel commits multiple homicides at routine intervals, many of which are individually unremarkable. While some players may feel that the quantity of violence in the game adds to the sense of a hellish dystopia and Joel’s character, others may feel that the TV format is more effective at exploring the story’s darker themes.
Still, the realism of the digital Joel and Ellie may help us to empathize with them. For Cairns, realistic visuals can provide a kind of “shortcut” to empathy. “They’re able to show you the thing that they’re talking about, so you get into that world quicker.” Even so, it doesn’t necessarily follow that more detail equals more impact, he points out, since we can be equally shocked or disturbed by scenes described in books, while games such as Papers, Please, which eschew realistic graphics in favor of stylized retro visuals, are no less capable of forcing us to face harrowing consequences and feel something in the process. Visual realism is merely one possible avenue here.
Where realism may matter more is when games emulate real-world violence that connects with social issues. The trailer for Unrecord has also generated discussion due to its subject matter—a lone armed police raid on an apparent criminal hideout shown from the perspective of the officer’s body cam. The choice made by French studio DRAMA to use the bodycam view is more than an aesthetic one. Certainly in the US, but elsewhere too, bodycams link to questions of police accountability, especially in regard to racist violence and unjust harassment. Or, conversely, to questions about invasive surveillance.
What DRAMA intends to do with the narrative is so far unclear, although the developer put out a statement in response to viewers’ concerns. “The public generally trusts film, series, and novel writers on the intelligence of the point of view when it comes to detective, gangster, or police stories,” the statement reads. “Why not for a video game?”
It’s a fair point—if we are to treat games as we treat other “serious” media, we can’t assume that Unrecord will be crass and exploitative. Yet at the same time, there’s room for skepticism here. “The game will obviously avoid any undesirable topics such as discrimination, racism, violence against women and minorities,” the statement also says. Why “obviously”? If anything, bodycam footage of a police officer dealing with suspected criminals seems like fertile ground to critically and intelligently explore such topics. At the very least, we can hope that the bodycam recording has a bearing on your choices, working as a kind of moral and legal eye on your activities.
Does it really matter how Unrecord represents police violence, though? Well, potentially, yes. “What it does is, it plays into or reinforces existing stereotypes and prejudices,” Cairns says of the ideological framing of violence. It’s a comment that seems especially relevant in the context of the forthcoming Six Days in Fallujah, which, according to developer Highwire Games, “recreates true stories of Marines, soldiers, and Iraqi civilians” during the Second Battle of Fallujah in 2004. In one sense, then, it’s a “realistic” perspective of the conflict, but nothing we’ve seen yet suggests that it will touch on the controversy surrounding US tactics in Fallujah, including the use of white phosphorous munitions in the conflict. In short, it seems likely that this portrayal will do more to reinforce official US Army accounts than question them or portray the complicated nuance of war that soldiers on the ground face, not to mention that of civilians caught in the crossfire.
Also, research suggests that the narrative motivations behind violence in games may have an effect on behavior. “Context matters,” Drummond says, citing a study published in the Journal of Experimental Psychology, in which he and his colleagues asked people to play a violent game with either a heroic or anti-heroic narrative. Afterward, they tested players’ aggression levels and found that while violence in itself didn’t affect post-game behavior, the narrative did make a difference. “Players who played an anti-heroic character aggressed slightly more post-game than players who played a heroic character,” he says. “The context in which violence is presented does appear to be important in terms of its psychological effects.”
One more aspect of gaming that could be more impactful on depictions of violence than visual fidelity, meanwhile, is the rapidly increasing quality of AI systems. In most cases, NPCs are still little more than symbolic representations of people that fulfill a particular purpose—combat adversaries, merchants, and so on. But with large language models such as ChatGPT, relationships could be much more natural and textured. “If [a character] could realistically beg for mercy in a different way every single time you try and do something to it,” Cairns says, “that could be quite traumatic.”
At that point, enacting virtual violence wanders into some gray areas, Cairns believes. “If it’s essentially killing what feels like real characters, then I think you’re crossing a line,” he says. “I think there’s got to be a point that you don’t want to go past, that we don’t consider acceptable in society.” But by the same token, advanced AI models could also help games to explore the real impact of violence on people’s lives, he adds. As ever, then, it’s not only the level of realism in game violence that matters but how it’s used and presented. As the tech advances, now may be the time for designers to consider that with great computing power comes great responsibility.
