The Record (Bergen County, NJ)


by Howard Fienberg
September 29, 2000

The number of grotesquely violent video games has been increasing, letting young players slice the limbs or decapitate the heads of the "bad guys," fill their opponents full of bullets, or even incinerate them with portable nuclear weapons.

Long concerned over the relationship between media violence and the real thing, we have been riveted by Columbine and other recent instances of youth violence. Joe Lieberman's vice-presidential candidacy highlights his campaign against violent media, which he feels is a trigger for a violent society. And Congressional hearings this week on Hollywood fare are bringing the subject to the front pages.

Might video games be a gateway device, driving the average teenager to turn his or her virtual world into actual violence? Don't make any presumptions just yet. Research has so far failed to identify any concrete link, and current studies indicate we may never get the evidence locked and loaded.

Several dated studies that connected virtual and real-world violence failed to control for differences between violent and non-violent games. Could mundane video games like Pong inspire violence the same as first-person shoot-em-up games like Doom? Countless other experiments failed to control for important factors that potentially obscure what arouses a kid when he or she plays a violent video game, including violence, general gore, speed, bright colors, and difficulty level.

Better conceived than most, a 1987 study from the journal Psychology specifically selected games similar in dimensions of excitement, difficulty, and enjoyment, yet could find no effect from violence. Professors Craig Anderson and Karen Dill admit that "there is presently no empirical evidence on whether playing a violent video game increases accessibility of aggressive thoughts."

So in April, Anderson and Dill published their most recent research on violent video games in the Journal of Personality and Social Psychology. They ran two studies of university students. The first was a vague survey of subjects' favorite video games and their aggressive and delinquent habits.

Its conclusion that playing violent video games was linked to real world violent behavior and delinquency was based on correlation; there was no evidence of causation. It links a personality type to a pastime, so it could be that people who are violent and criminal play violent video games. Unfortunately, it may not be the case that those video games cause them to be violent.

The second study was more useful. One group played the violent (but rather outdated) Wolfenstein 3-D, while another group played the non-violent Myst. Both groups were tested for aggression afterwards.

The first test measured the speed with which subjects could repeat aggressive words that were flashed on a screen in front of them. But psychologist Guy Cumberpatch, director of The Communications Research Group, disregards this measurement, since it is hard to tell if the violent video game speeded up responses, as Anderson and Dill claim, or if the non-violent game slowed responses.

The second test was held after a similar game play episode a week later, and checked subjects reaction times to sound. Those that reacted fastest were then given the opportunity to blast an opponent (which they thought was another human, but was actually a computer) with a loud noise and those that were too slow were subjected to a noise themselves.

The researchers found that those that had played the violent game blasted their opponent longer and louder than those that had played the non-violent game. But the difference was actually minimal. The blasts delivered by subjects who had played violent games were longer, by all of 2 percent, and the average blast for all the students was about half a second, far too short for reasonable analysis.

So Anderson and Dill's conclusion, that "playing a violent video game increased the aggressiveness of participants"is not very convincing. They found an association, but are light years away from establishing causation.

If violent video games are part of the problem, they are not obviously the discernable cause of that problem. These games are at times more popular abroad, but youth violence is far more prevalent here than in other countries. While this does not preclude a link, it does usefully highlight that the issue is far more complex than we might expect.

A measure of "aggression" may not be a useful proxy for real-world violence which, while troubling, is relatively rare.

The jury has yet to conclude that violent video games create violent offenders, and the difficulties facing research to date show no signs of disappearing. Researchers first needed to understand that not all video games are alike. Now, they need to realize that not all children are alike.

While it is quite possible that fixation on violent video games may herald problems for a disturbed child, the average child probably just plays his games and grows up fine.

Howard Fienberg is research analyst with the Statistical Assessment Service, a non-profit, non-partisan think tank in Washington, D.C.

return to Howard Fienberg's page