The Sandbox

Here's a piece I wrote for Electronic Gaming Monthly about the studies:

Off Target
David Kushner, Electronic Gaming Monthly, November 2006

We can assume two things about you if you’re reading this magazine: You don’t think playing violent videogames can make someone go aggro in real life, and you haven’t authored any studies linking violent games to violent behavior. But the people who do believe and have authored such studies have gotten a lot of play lately in the mainstream media—and they’re putting the future of your favorite pastime at risk.

Following the April 16 Virginia Tech shootings, the Washington Post reported online that the killer had a history of playing the PC squad-based multiplayer shooter Counter-Strike. By the time the paper took down the reference from its website the next day (due, the writer later said, to a necessary update), it was too late. Ubiquitous antigame crusader Jack Thompson raised the specter on CNN. Dr. Phil played the blame game on Larry King Live. “The mass murderers of tomorrow are the children of today that are being programmed with this massive violence overdose,” he said.

Then on April 26, the Federal Communications Commission weighed in with its report, three years in the making, on the impact of media violence (particularly television violence) on kids. It suggests that Congress can step in to protect kids from harm by regulating violence on TV without violating the First Amendment. The thought of the Feds legislating videogames strikes many as dangerous. The American Civil Liberties Union calls it “political pandering.” Howard Stern calls Dr. Phil an a-hole. Once again, the debate that has run from Columbine to Blacksburg continues to rage. And when it does, each side looks to the same place to buttress their arguments: scientific research on the effects of violent videogames. But with sensational media and political distortion in the way, getting to the truth of the research is the trickiest game of all.

Anger management

At the end of the day, scientists—including those behind the studies cited in the FCC report—still aren’t sure if playing violent games leads to real-life violence at all. “The research doesn’t support the notion that [playing violent games] leads to aggression,” says Dr. Jonathan Freedman, a psychologist from the University of Toronto. “It doesn’t even deal with the question of whether it leads to criminal violent behavior or real violence. At most, it addresses the question of whether it leads to aggression, which I don’t think it does.”

One of the problems with the studies is how the term “aggression” is defined. “The missing element is that most of these studies, if you look at them just a little bit critically, don’t really measure what a lot of people purport they’re measuring, and people don’t understand how they fall short,” says sociologist Dr. Karen Sternheimer of the University of Southern California and author of Kids These Days: Facts and Fictions About Today’s Youth. While the general public equates aggression with violent behavior, actual violent behavior has never been measured—for obvious reasons. “We can’t have people assault, rape, or murder someone” in the lab, says Dr. Brad Bushman, a University of Michigan psychologist who studies the effects of media violence. Instead, researchers are left to measure innocuous examples of so-called aggressive behavior—behavior that doesn’t remotely resemble criminally violent activity. This has ranged from having subjects punch an inflatable Bozo doll to, more commonly, blast opponents with a loud noise.

Even Dr. Karen Dill, who with Dr. Craig Anderson coauthored one of the most-cited studies—2000’s “Video Games and Aggressive Thoughts, Feelings, and Behavior in the Laboratory and in Life”—admits “hearing the noise is not harmful.” Nevertheless, the report opens with an allusion to Columbine and purports that “one possible contributing factor is violent games.” To many, that’s an egregious leap. “Pressing a button that delivers a short burst of loud noise is pretty remote from real aggression,” Freedman notes.

Old data

But it’s not just the measures of aggression that are questionable—it’s the means through which participant reactions are elicited in the first place. Reading the fine print in the Dill and Anderson study, for example, reveals that the researchers used outdated, mismatched games and required an absurdly brief amount of actual playtime from the subjects. The researchers compared the response to people playing two games released in the early 1990s: Wolfenstein 3D, the first first-person shooter, and the puzzle adventure Myst. The disparity between the game styles raises questions about the results. Though the goal of the study is to explore the effect of violent games on aggression, a shooter is sure to elicit more aggressive behavior than a puzzle game. It’s like comparing apples to hand grenades. Wouldn’t it have been better to compare two action games—one with violence and one without?

The study required 32 undergrads to play the games for 15 minutes each. They were then given the opportunity to send a noise blast to an opponent—often just a computer proxy—after they finished the game. “You can’t study people for 20 minutes and know what’s going to happen to people in society 10 years later,” says Dr. Dmitri Williams of the University of Illinois at Urbana-Champaign. Williams recently authored one of the first long-term studies, in which he observed players of the online PC role-playing-game Asheron’s Call for more than 56 hours in a period of a month. His results? “I found no evidence of increased aggression or aggressive attitudes,” he says.

Dr. Patrick Markey, a psychology professor at Villanova University, decided to take another perspective: studying what role a person’s anger level before playing a game has on the aggressive behavior coming out. And Markey, unlike some of his colleagues, actually uses games played in the last decade. The 167 students who participated played games such as Doom 3 and Project Gotham Racing. His conclusion: The people who had previously filled out questionnaires reflecting an even-keel personality were less aggro after playing a violent game. Those who had a more aggressive disposition were more susceptible to these heightened emotions.

While some could conclude in broad strokes that games cause aggression, the nuances tell another story, Markey notes. “The general research shows there is an effect of violent games on aggression, but what gets lost is [that] this effect isn’t that big,” he says. And, of course, videogames aren’t the only pastimes that could lead to aggression: dodgeball, paintball, and a bad beat in Texas Hold ’Em can heighten arousal, too. Dr. Vincent Mathews, a radiologist at Indiana University who has studied the brain’s response to violent videogames, suggests that the effects of these other activities would be comparable. “I would think that paintball or dodgeball would show similar results,” he says. But no one is calling for these games to be banned.

Popular science

Critics of violent games cite the studies as further proof that media violence leads to murder. As Thompson wrote in March 2007, “The American Psychological Association [APA] in August 2005 found a clear causal link between violent games and teen aggression.” But as political watchdog site GamePolitics.com astutely reported, Dr. Elizabeth Carll, who co-chaired the study, wanted to make clear that “the resolution did not state that there was a direct causal link to an increase in teen violence as a result of playing videogames. Rather, [it stated] an increase in aggressive behavior, aggressive thoughts, angry feelings, and a decrease in helpful behavior as a result of playing violent videogames.”

If no one has said there’s a causal link between games and real-life violence, why does it keep making headlines, and why do these studies get cited so much? “The [American Psychological Association] is a political organization...and they do what is politically expedient like any other group,” says Dr. Christopher J. Ferguson of Texas A&M International University’s Department of Behavioral, Applied Sciences and Criminal Justice. Ferguson recently released a study named, with typically academic wordiness, “Evidence for publication bias in videogame violence effects literature: A meta-analytic review.” In it, he finds what he calls “a systematic bias for hot-button issues” that results in overstatements and misleading results.

The authors of the reports bristle when their research is challenged. Dill, after agreeing to be interviewed for this story, later e-mailed to request that her interview not be used because of what she perceived to be an effort to “push the tired ‘party line’ that the research is wrong.” Her colleague, Anderson, declined entirely, saying an interview would be “pointless.”

But it’s not just their research that’s being challenged—it’s the manner in which the findings are presented. “From the present body of literature, there’s nothing that supports a relationship between violent videogame playing and aggression—not correlational or causal,” Ferguson says. “The moral of the story is that scientists ought to be using much more measured tones in discussing what has become a political issue rather than giving in to the urge to engage in hyperbole.” In other words, violent games sell—not to kids, but to the general public at large. Like Elvis in the 50s, or Dungeons & Dragons in the 1980s, videogames are still viewed as the dangerous scourge of youth culture. In the face of awful, inexplicable tragedies, media violence is an easy target.

Truth be told

What’s lost to the game-violence critics and public is a dose of reality, not only about the truth of the results but the context. “I don’t think they understand the way the media are used in daily life enough,” Williams says of the researchers. “They tend to focus more on lab research and ignore long-term research. People in the psychology community are less likely to pay attention to the social context of media use.” But others are. The British Board of Film Classification conducted a survey that found that “the violence helps make the play exhilaratingly out of reach of ordinary life.... Gamers seem not to lose awareness that they are playing a game and do not mistake the game for real life.”

And considered in light of recent youth crime statistics, all the noise blasts don’t pass the muster of common sense. In 2005, for example, just 12 percent of the videogames sold were violent enough to bear an M-rating by the Entertainment Software Ratings Board, the industry’s voluntary ratings group. At the same time, youth crime is dropping precipitously. The number of kids under 17 who committed murder fell 65 percent between 1993 and 2004. “If this was affecting all kids in a bad way we’d see something,” argues Dr. Cheryl Olson, professor of psychiatry at the Harvard Medical School’s Center for Mental Health and Media.

Even the surgeon general’s youth-violence report, which the FCC cites in its recent findings, couldn’t find a convincing link. “Taken together, findings to date suggest that media violence has a relatively small impact on violence,” the surgeon general reported. And the specific inferences about game violence were even less swaying. “The overall effect size for both randomized and correlational studies was small for physical aggression and moderate for aggressive thinking...,” the surgeon general found. “The impact of videogames on violent behavior has yet to be determined.”

So what are we left with? A possible link between violent media and loosely defined “aggressive behavior” (noise blasts, clown-doll punching, and so on) but no evidence that playing violent games actually causes violent—let alone criminal—actions in real life. “It’s time to move beyond blanket condemnations and frightening anecdotes and focus on developing targeted educational and policy interventions based on solid data,” Olson suggested. “As with the entertainment of earlier generations, we may look back on today’s games with nostalgia, and our grandchildren may wonder what the fuss was about.”