Early last year, 70 French university students sat down in a room. A group of scientists told the students they would be participating in a study to measure the effects of video game brightness on visual perception, and that they would each be paid €10 ($13) a day for their efforts.
The students split up into two groups of 35, and each group was randomly assigned to play a violent or non-violent game 20 minutes a day for three days. The violent games: Condemned 2, Call of Duty 4 and The Club. The non-violent games: S3K Superbike, Dirt 2 and Pure.
After each gaming session, the students were asked to write an ending to one of a few randomised stories. In one of the stories, for example, a driver crashes into the main character's car, causing a bunch of damage, and then the main character gets out of his car and approaches the other driver. Students had to fill in what would happen next.
Then, each student was told that they would compete in a computer game in which they had to respond to a visual cue faster than an opponent of the same gender. The loser, the students were told, would receive a heinous noise blast that sounded like a combination of fingernails scratching a chalkboard, dentist drills, and ambulance sirens. Each student could determine the intensity and duration of the noise that their opponent would have to suffer through.
The results of this experiment -- conducted by Ohio State University professor Brad Bushman not to measure brightness, as they had told the students, but to examine the connection between violent video games and aggressive behaviour -- were conclusive. The people who played violent video games were more likely to write more aggressive stories and dish out higher, more unpleasant noise. Violent video games, Bushman and his colleagues concluded, have a direct causal effect on aggression.
In other words, playing Call of Duty makes you want to fight.
This week, US President Barack Obama asked Congress to dedicate $US10 million toward studying the effects of violent media -- including games, which he singled out during a speech Wednesday morning. In the wake of last month's tragic shooting at Sandy Hook elementary school -- and the revelation that killer Adam Lanza had enjoyed shooter games like Call of Duty -- violent video games have again become a significant topic in national conversation. And as part of a bigger plan to fight gun violence, Obama wants to see more studies on how games affect our behaviour.
But do we really need more research? What about the studies that have already been done? Have researchers found any links between video games and violence? Will violent video games really make kids more aggressive? Or is this all just a massive waste of time and money?
Over the past few weeks, Kotaku has gone through dozens of studies and spoken to multiple leading researchers in the field of violent media. While there are no documented scientific links between video games and criminal violence, the question of whether violent video games lead to aggression has been hotly debated. (That distinction between criminal violence and aggression is critical. Science has yet to show any links between video games and violence, but violent games may have a more subtle effect on children: for example, they could make a child more inclined to bully or spread rumours about his peers.)
Some scientists, like Bushman, have concluded that yes, playing violent video games will make children more aggressive. Others argue that current studies are faulty and inconclusive.
It's a debate that has been going on for over 25 years. And it shows no signs of stopping.
Violence And Video Games
From the spine-ripping of Mortal Kombat to the increasingly-realistic war zones of Call of Duty, video game violence has been criticised and scrutinised for decades now. You've probably heard the theories, maybe even voiced them. Violent video games desensitise kids to real violence. They make kids more unruly, more likely to fight one another. Playing games like Halo makes us more aggressive.
This conversation bubbles up most frequently after a national tragedy involving guns. After the 1999 Columbine shooting, for example, politicians and reporters were fond of emphasising the connection between the killers and their favourite game, the first-person shooter Doom. As University of Toledo associate professor of psychology Jeanne Funk told the Los Angeles Times in one 1999 article: "We found signs that children who enjoy [violent] games can lose the emotional cues that trigger empathy."
Then there were the connections between the Virginia Tech shooter and Counter-Strike. And the Norwegian mass murderer who loved World of Warcraft. It seems like every time there's a tragic shooting, the conversation inevitably turns to violent video games.
WHY NOW? Iowa State professor Doug Gentile hates the fact that this conversation only comes up in the wake of tragedies like the Sandy Hook shooting. "Why is it the only time we talk about media violence is when there's a horrible event?" he said. "Once we have a horrible tragedy like this, it really distorts the way we think about the issue... we have what I call a culprit mentality. ‘What's the cause of this?' Well, it's never the cause. There's never one reason for anything like this. There's never one reason. Humans are complex."
For gamers, this is all tired ground. Longtime video game fans are sick of defending their hobby from pundits like Jack Thompson, the now-disbarred lawyer who infamously railed against games like Grand Theft Auto and Bully, and Carole Lieberman, a psychiatrist who made headlines in early 2011 when she said that video games led to rape. (There is no published scientific evidence that video games lead to rape.)
But what about the more reasonable claims? Though scientists have not found any links between video games and violence, there is a heated debate over another, more relevant question: Do violent video games make us more aggressive?
To answer that question, we have to look at the science, which, oddly enough, hasn't changed all that much over the past few decades. The first major violent video game study took place in 1984. It surveyed 250 high school students and quizzed them about their game-playing habits and aggressive behaviour with questions like Somebody picks a fight with you on the way home from school. What would you do?" The researchers eventually concluded that physical aggression was linked to arcade video games, writing: "The data indicate that video game playing is neither the menace that many of its critics have portrayed it to be, nor necessarily without possible negative consequences."
Since then there have been over 100 studies on the effects of game violence and aggression, according to Chris Ferguson, a professor at Texas A&M and one leading researcher in the field of media violence. Some of these studies put college students in a room and ask them to play certain games; other studies measure children's behaviour over time, through regular reports that their parents would submit every month or so. Some studies test a student's aggression using noise tests and story stems, like the study mentioned above. Others use more unusual metrics, like the hot sauce test. (More on that later.)
For the layperson, trying to parse all of these studies can be rather intimidating, so scientists often publish what are called meta-analyses -- studies that look at a wide range of experiments and try to draw conclusions from the patterns they find. In the world of violent video games, there have been two major meta-analyses, done by two different groups of people who examined the same set of data.
They came to totally different conclusions.
A Tale Of Two Studies
If you spend a significant amount of time reading research on violent video games, you'll notice some names that pop up again and again: Brad Bushman, Craig Anderson, Chris Ferguson. You'll also notice that those people tend to fall into two camps.
On one side of the argument are Bushman, Anderson, and several other scientists who say there's a definitive causal link between games and aggressive behaviour. Violent video games, this camp would argue, make people more aggressive.
"On average, the research shows that exposure to violent video games increases aggressive thoughts, it increases angry feelings, it increases physiological arousal such as heart rate and blood pressure, which may explain why it also increases aggressive behaviour," Bushman told me in a phone interview. "It decreases helping behaviour and it decreases feelings of empathy for others and the effects occur for males and females regardless of their age and regardless of where they live in the world."
Then there's the other side of the argument, supported by Chris Ferguson, Cheryl Olsen, and a handful of other researchers. The evidence, this camp says, just isn't conclusive.
"My impression is at this point the research is certainly inconsistent," Ferguson told me. "I think anybody who tells you that there's any kind of consistency to the aggression research is lying to you, quite frankly... There's no consistency in the aggression literature, and my impression is that at this point it is not strong enough to draw any kind of causal, or even really correlational links between video game violence and aggression, even, no matter how weakly we may define aggression."
So scientists are divided, to say the least. And while both Bushman and Ferguson explicitly told me that they have a great deal of respect and admiration for one another, their conclusions are mutually opposed.
Let's take a look at each argument.
The Case For A Link
"It looks like a pretty clear link," said Doug Gentile, a leading researcher in media violence who spoke to me on the phone last week. "Kids who play more violent video games -- it changes their attitudes and their beliefs about aggression. It does desensitise them. It certainly hypes up aggressive feeling in the short-term. In the long-term it probably links aggression with fun, which is a really weird idea. Or aggression and relaxation, another weird idea."
Gentile, a professor at Iowa State who has worked on a number of studies and said he's read many more, doesn't see inconsistency as a problem.
"Just because one study comes out showing no effects doesn't mean the other ones were wrong," he said. "It's really easy to wipe out an effect. It's hard to have an effect show up over and over again if it's not a real one."
Brad Bushman agrees. "I could do the same exact study 50 times and get different results every time," he told me. "There's random variability in every scientific study, and some studies find null effects. The problem with null effects is that they're very difficult to explain... It's much easier to explain an effect than no effect at all, because there are many different explanations for a null effect."
In 2010, Bushman ran a meta-analysis called "VVG Effects on Aggression, Empathy, and Prosocial behaviour in Western and Eastern Countries: A Meta-Analytic Review." He and some colleagues studied results from something like 130,000 participants, concluding that there is indeed a link between violent video games and aggression.
The meta-analysis looks at dozens of studies done both in the U.S. and Japan, many of which follow the same pattern: First, a group of people is measured and tested. Scientists record their heart rates, excitement, skin conductivity (sweatiness) and other psychological factors. Next, a portion of the group spends time playing a violent video game, and another portion spends time playing a non-violent video game. They are then tested in various ways. Some studies show images of real-life violence and measure subjects' brainwaves in order to see whether they're more desensitised to horrifying pictures. Other studies use the loud noise test or other direct measures of aggression.
So the science seems to show a link between games and aggression, but what about violence? There's no way of knowing, Bushman said. Violence can't even be tested.
"We can't give our participants knives and guns and see what they do with them," he said. "It's not ethical to do that. But we can use ethical measures in which they can harm another person physically or otherwise and those measures consistently show that violent game players are more aggressive than non-violent game players.
"Are they more likely to stab someone? I dunno. Are they more likely to shoot somebody? I don't know. Are they more likely to rape someone? Beats me. Those are very rare events and we can't study them ethically, so I don't know what the link is between playing violent video games and violent criminal behaviour. But we know that there is a link between playing violent video games and more common forms of aggressive behaviour -- such as getting in fights."
The Case Against A Link
Chris Ferguson doesn't just think there's no proven link between violent video games and aggression: he thinks today's studies are totally bogus.
"There are over 100 studies at this point that in some way or another tap into video game violence and aggression," Ferguson said. "Most of them are horrible."
For one, the meta-analysis conducted by Bushman and crew suffered from publication bias, Ferguson said. Many studies that show null effects remain unpublished, as researchers tend to assume that in those studies, something must have gone wrong.
One of Ferguson's studies, for example, split university students into different randomised groups. Some played the graphic shooter Medal of Honor; others played the non-violent adventure game Myst III. "No link, either causal
or correlational, was found between violent-video-game playing and aggressive or violent
acts," he wrote.
Ferguson, who met last week with vice president Joe Biden and several leaders in the gaming industry to talk about violent video games, thinks there are three main flaws with today's research. The first: many of the studies look at college students, not children.
"Of course most of these college students probably have heard theories about media violence and aggression, ‘cause they're in college and taking these classes," Ferguson told me. "So a typical experiment is they show you a violent video game and ask you to be aggressive one way or another, and probably a typical college student can draw that link of what they're supposed to do, basically."
College students are more likely to show evidence of aggression than kids, Ferguson said. "It's kind of the opposite of what we'd expect, developmentally and the reason for that probably is because these college students are guessing what they're supposed to do and doing it, in order to get their extra credit."
The second major flaw with current studies, in Ferguson's view, is that measures for testing aggression are not ideal.
"They're kind of like filling in the missing letters of words, so if you spell explode rather than explore, that indicates you're being aggressive," he said. "Or you may be giving people little bursts of white noise. These measures that are being used are not very effective at getting at even minor acts of aggression."
The third major flaw is called the methodological flexibility problem, Ferguson said. Some of the testing measures used in these scientific studies are so flexible that researchers can pick and choose outcomes that fit the hypotheses they want to achieve.
As an example, Ferguson pointed to the noise test (used in the above French university study, among others). He cited an experiment done by a German researcher named Malte Elson, who found that it's possible to draw a number of different outcomes from that particular measure.
"From the same noise burst test, you can either show that video games increase aggression, decrease aggression, or have no effect at all," Ferguson said. "So the concern is that researchers that have a particular belief system are just picking outcomes -- in good faith, nobody's saying that they're doing it on purpose or lying -- that best fit their hypotheses. And of course that's a big problem."
US legislation has also supported Ferguson's argument. In 2011, a Supreme Court case struck down a law that would have made it a crime for stores to sell violent games to kids. After a well-publicised battle, the Court determined that there was no conclusive link between video games and aggression, writing that "most of the [violent game] studies suffer from significant, admitted flaws in methodology."
What Is 'Aggression'?
Both sides seem to agree on one thing: the majority of published studies do show some sort of connection between video games and aggression. Both Ferguson and Bushman's meta-analyses show an average aggression effect size around 0.15, a number obtained by taking two variables -- violent video game play and aggression symptoms -- and entering them into a complex mathematical equation.
By taking that number and squaring it, you can find the percentage of increase in a certain behaviour or effect. So if you square 0.15, you get 0.0225 -- or an aggression increase of roughly 2%.
But how do you quantify aggression? Even scientists disagree on the answer to that question. In one of Ferguson's studies, for example, he found an effect result of 0.06 -- or a percentage increase of less than half a per cent.
"Let's imagine you played a violent video game and it made you one half of a per cent more aggressive -- would you notice that? I don't think you would," Ferguson told me. "To put it into context, if tomorrow you're one half of a per cent more happy than you are today, what does that really mean? It's a very tiny effect... If my son was one-half a per cent more aggressive today than he was yesterday, I'd never notice that."
But Gentile disagrees, saying Ferguson cares more about the connections to violent crime. A small aggression increase might affect kids in more mundane ways.
"What I care about is everyday aggression," Gentile told me. "What's that? Well, walk into a junior high school and look at how the kids treat each other. They're unkind to each other. They say mean things. They spread rumours about each other. They ostracise each other. They give each other the cold shoulder. They are verbally aggressive towards each other, and sometimes maybe even they'll hit. That's like the most extreme thing you're gonna find, right? Those are the levels of aggression that kids actually have in their lives.
"Not just kids -- adults do too. If you think about the aggression in your life, that's what it is: you think of your friends or your girlfriend, you might get really angry and say something unkind. That is true aggression, and it does hurt, right? So what I'm talking about is serious as far as I'm concerned in terms of peoples' lives. Serious, real-world aggression. And when we look at that level, well the effect seems to be there, including in Ferguson's study."
Aggression -- which can be broadly defined as any form of hostile behaviour -- is measured in several ways: noise tests, story stems, self-reported surveys. And then there's hot sauce.
The hot sauce test, one of the oddest things we found in violent video game research, is similar to the obnoxious noise test: a participant is told that he or she must give hot sauce to a partner (who doesn't actually exist) as a way of punishment. The participant gets to decide just how hot the hot sauce will be, and how much to give.
Seems strange, right? Bushman told me it's actually a well-tested way to prove aggression, pointing to the infamous "hot sauce mom" as just one example of a person using spicy food as an act of aggression.
But what sort of conclusions can we really draw from a student choosing to dole out spicy hot sauce? Can we really link verbal or physical abuse to a test that seems so strange? It's measures like this -- and really, the ambiguity of "aggression" as a psychological concept -- that have made professors like Chris Ferguson sceptical of today's research, even when the evidence seems conclusive.
Here's another question: have researchers successfully isolated violence as the problem here? Or is there another factor at play here?
That's what Paul Adachi has been asking. Adachi, a PhD student at Brock University, recently published a longitudinal study that he conducted with his professor, Teena Willoughby. They examined 1,492 adolescents over four years, monitoring the kids' video game habits and measuring how much time they each spent playing different types of games: sports games, racing games, shooters. The kids would then fill out confidential reports about their behaviour, answering questions like "How frequently in the past six months have you kicked or hit someone?"
Willoughby and Adachi tracked violent competitive games (ex: Mortal Kombat Vs. DC Universe), violent non-competitive games (Left 4 Dead 2), non-violent competitive games (Fuel), and non-violent non-competitive games (Marble Blast Ultra). What they found was fascinating: it wasn't violence that triggered aggression; it was competition.
"We found that playing more hours a day of the two types of competitive games did predict aggression over time," Adachi told me over the phone. "Whereas playing non-violent, non competitive games did not. So that really gets at the idea that, well, it may not be the violence, it may be the competition in games that is responsible for a link between video games and aggression."
Competition is just one factor that must be considered when studying the effects of violent video games on aggression. What about gender? Boys are more aggressive than girls. How about home life? Income level? Previous cases of bullying or being bullied?
In the studies Kotaku looked at, some of these factors were controlled. Some were not.
"From what I've seen, there's only been two studies where they really controlled well for competition in the games," Adachi said. "But there's different factors that weren't controlled for that could also be related to aggression, such as how difficult the game was or how fast-paced... It's a hotly debated topic."
Are Games Worse Than Movies?
If you ask an average parent what's worse for their kids, a violent video game or a violent movie, they'll probably pick the video game. Makes sense, right? After all, controlling a guy with a gun would theoretically affect your brain more than just watching a guy with a gun.
But believe it or not, we couldn't find a single study that compares the effects of violent games to the effects of violent movies or TV shows. And none of the researchers I spoke with knew of any.
There was one study, conducted in 2008 by a scientist named Hanneke Polman, that brought 56 kids (28 boys, 28 girls) into a room, divided them into groups of three kids each, and had each group sit in front of a video game. One child would play a non-violent game, one child would play a violent game, and one child would watch that same violent game. Later, they were all tested. Polman and his colleagues found that players of the violent game were significantly more aggressive -- at least in the short-term -- than people who just watched it.
"It's a beautiful study because they saw exactly the same violent images," Brad Bushman told me. "Which would be not the case if one watched a movie and another played a video game. They saw exactly the same violent images, but the players were more aggressive than the watchers. So we need more studies like that."
Where's The Money?
As in all scientific studies, it's important to look at the politics behind violent video game research. One factor worth considering: who funds all of these studies?
"Some scholars have taken research funding from advocacy groups, which is just as bad as taking research funding from the video game industry, as far as I'm concerned," said Ferguson, pointing to groups like the now-defunct National Institute on Media and the Family and the centre For Successful Parenting, a family-values group that sets out to find the negative effects in violent media.
"It's something we need to stop on both sides."
Just a couple of years ago, when a study came out claiming that violent video games have a long-term effect on brains, Rock Paper Shotgun's John Walker did some legwork and found that the centre for Successful Parenting had funded the study. While that certainly doesn't make the findings any less valid, it calls into question those scientists' motives and hypotheses.
And what of the video game industry? I reached out to Dan Hewitt, a representative for the Entertainment Software Association (the group that helps regulate and represent gaming companies), to ask.
"ESA hasn't funded any research in any way," Hewitt told me in an e-mail. "Everything that's out there and that we talk about is completely free from any ESA influence or financing."
What We Should Do?
You don't need a doctorate to know that the human brain is a complex machine, and that nothing about our behaviour is predictable. There's nothing exact about social science: different people get aggressive for different reasons, and there are hundreds of different factors that could contribute to that.
But it's hard to argue with Obama's assertion that we need more research into the effects of violent video games. How could it be a bad thing?
Whether you believe that the link between violent video games and aggression is clear or you think the science is too faulty to mean anything -- and there are strong cases on both sides -- it's hard to argue that more research is unnecessary.
So maybe the data speaks for itself: maybe there is a clear link between video games and aggression.
"No researcher I know would say violence in the media is the only risk factor for aggression or violence or that it's the most important factor," said Bushman. "It's usually a cumulation of factors. But it's one factor that we can do something about."
Or maybe Chris Ferguson is right, and today's research is too inconclusive to determine any causal links. It certainly can't hurt to be more sceptical about what you see in the media.
"If what people take away from this is to be more sceptical about statements that are made by scholars on all sides of any debate, whether it's video game violence or something else, that's a wonderful thing," Ferguson said. "Because science is a human endeavour. The more someone tells me that they're absolutely objective, the less I believe they are. So people need to fact-check things. They need to understand that science is a human endeavour. Science is easily damaged by politics and personal opinion."
Holden Miller contributed reporting to this story.
Want to read some violent video game studies on your own? Here are a few: