Don't Blame the Games

In the wake of the recent school shooting at Marjory Stoneman Douglas High School, policymakers have placed increased emphasis on violent video games as a probable cause of gun violence. President Donald Trump, for example, is scheduled to meet with representatives from the video game industry to discuss the issue with them. The picture that the academic literature paints, however, is starkly different. Far from conclusively proving that violent video games cause school shootings, the state of the art in violent video game research instead demonstrates that no link can be confidently be made between these things.

This wasn't always the case. In the late 1990s and 2000s, it seemed increasingly certain that playing violent video games led to violent behavior among both children and adults. A series of landmark studies showed clear correlations over several thousand individuals: The more someone played violent video games, the more likely they would be to display antisocial behaviour, and the more likely they would be to commit acts of violence.

However, academics knew that just looking at correlations was not sufficient to prove a causal link between video games and violence. In order to do this, controlled experimental research was necessary. Early experiments did indeed suggest that the violence in video games led to violent behavior.

People who were exposed to violent video games in a laboratory setting showed increased amounts of aggressive thoughts and behaviors as opposed to individuals who had instead played nonviolent games. These effects led to the popularisation of an influential theory: playing violent games would temporarily make people more likely to commit violent actions. Over time, this effect would become reinforced, leading to players becoming permanently more violent.

In recent years, however, a growing number of academics have voiced concerns about the validity of this perspective. Correlations which confidently proclaimed that gaming led to violent behavior were shown to likely be the spurious by-product of factors such as family violence.

Follow-up studies which attempted to look at correlations in a more rigorous fashion found a relationship between violent video game play and violence that was close to zero percent. Academics noted that violent video game usage had risen dramatically in recent years, but youth violence had not risen at a similar rate – it had, in fact, declined. Accusations of publication bias in the academic literature abounded.

Most damningly, serious flaws were found in the well-regarded laboratory experiments which were supposed to conclusively demonstrate that playing violent video games led to violent behaviour. These studies hinged on the idea that one group of people would be given a violent game, whilst others would be given a nonviolent game. After play, measurements of factors like aggressive thoughts would be taken, allowing researchers to see what the violence in games did to people in a more fine-tuned way than just looking at correlations in society.

On further scrutiny, it was revealed that the games used in these experiments varied dramatically. Typically, the violent games that researchers used were significantly more difficult and complex than the nonviolent games. Thus, increases in aggression amongst people who had played the violent games might not be due to violent content at all – it might simply be because these games were frustratingly difficult.

Academics attempted to re-run these experiments and replicate their results. When the experiments were concluded, they returned 'null' results, indicating that no link between violent content and violent behaviour could be confidently found. More researchers came forward to argue that these null results had always existed – but had been suppressed by a field that was focused on the existence of a link between violent behaviour and violent games.

My own research tallies with this perspective. I have run video game experiments on over 4,000 participants, where they were tasked with playing first-person shooters, run-and-gun games, role-playing games and violent driving games. The experiments tested if there are any links between the violent content of video games and the priming of aggressive thoughts in adults in the most rigorous ways possible. Across all of these experiments I have not found any evidence that such a link exists.

This is the state of the art in violent video game research. This is a research community that is at war with itself. One group of academics stands firm on the position that violent video games are currently a legitimate source of danger to modern society. Another group of academics provides compelling arguments about why there is little convincing evidence that this is the case.

This group challenges the very validity of the research methods that their opponents rely on. They claim that the theory of video game violence is built on rotten foundations. There is currently no clear winner in sight in this contest for the future of video games. Both perspectives are well-represented in the academic literature. If you pick up a copy of a relevant journal (such as Computers in Human Behavior) and flick through the last few issues, you will almost certainly find researchers supporting both, radically opposed, perspectives. This degree of uncertainty and lack of consensus is extraordinary. It is not the norm in most fields of scientific research.

What does this mean for policy-makers? In essence, the research field is in tumult. It is currently unclear whether violent video games play any role at all in causing gun violence. There is currently no evidence in the literature which anyone can confidently use to causally link violent video game play and school shootings. Anyone who does so is almost certainly overstating their case. Policymakers should bear this in mind when making decisions.