Carmageddon 2 (source: Gamespot) is a gory racing game where players control drivers with names like “Max Damage” as they tear through city streets mowing down pedestrians and forcing competitors into bloody collisions. The game settings can be adjusted so that running down innocent bystanders actually increases a player’s point total. Surely, if there’s any video game that might raise a parent’s ire, Carmageddon 2 is one of them.
Studies have shown that violent video games are more likely than non-violent games to induce aggressive behavior, even after very short playing sessions. But more recent research (by Dmitri Williams and Marko Skoric) has suggested that violent game play does not always lead to aggression. In Asheron 2, Williams and Skoric note, players are not allowed to attack other humans, only monsters, and they must cooperate with other players to succeed. Perhaps by limiting violent actions and rewarding cooperation, Asheron 2 discourages aggressive behavior. Other studies have shown that people who watch violent television shows where the perpetrator is punished are less aggressive than those watching violent TV behavior that is rewarded.
Nicholas L. Carnagey and Craig A. Anderson wondered if punishment and reward are keys to whether violent video games lead to real-world aggressive behavior. Video games are different from TV, and players might become frustrated if they are punished for violence, leading to more real-world aggression. Noting that no study had previously controlled the punishment and rewarding of violence in video games, they designed a study to do just that.
Carnagey and Anderson utilized Carmageddon 2′s customizability to create three study groups: in one group, participants were rewarded with points for killing innocent pedestrians and smashing into opponents; in a second group, they were punished for it; in a third “non-violent” group, pedestrians were removed from the game entirely and computerized opponent cars were programmed to behave passively.
Participants played the game for just 20 minutes, but even in this short time, different game-playing behavior was observed. Players who were rewarded for violence typically killed 80 pedestrians; those who were punished killed fewer than a third as many in the same period of time. Next the researchers studied post-game aggression in three different ways. First they measured aggressive affect. This was done by giving the State Hostility Scale, in which participants rate their feelings in categories such as anger, hostility, and aggravation, on a simple 1-to-5 scale. For a second group of players, they used a Word Fragment Task, where participants complete as many words as possible in 5 minutes. For example, “K I _ _” could be completed as “kiss,” “kill,” “kick,” or “kilt.” Those who used more aggressive options in this task were rated as having more aggressive cognition.
Finally, a third group was tested for aggressive behavior. Participants were led to believe that they were competing against another participant playing a different game in an adjoining cubical (after the experiment, participants were debriefed and the true purpose of the study was explained). They were cleverly primed to dislike this “player” by being asked to write an essay expressing their opinion on a controversial issue—abortion. After they completed their essay, the experimenter took it and told them their competitor was going to “grade” the essay. Then the participants graded the fake competitor’s essay, which was chosen from two pre-prepared essays, so that participants always saw a paper expressing the opposite opinion from their own on the topic.
They played one of the three versions of Carmageddon for 20 minutes, and were given their “graded” essay, marked by the fictitious competitor with the lowest possible grade and indicating that “this is the worst essay I have ever read!”
Now, they were asked to perform a competitive reaction-time test. If a player won, he or she got to choose the “penalty” for their new nemesis (who they believed to be either a baby-killer or a woman-hater and also a poor judge of writing ability). The penalty was a painful noise played through headphones. Players chose before each test how loud the noise would be, and how long it would be played. In reality, the “test” was rigged so that the participant won 13 times and the non-existent “competitor” won 12 times, inflicting a noise blast of a pre-selected length and intensity. When the participant won, he or she got to get “revenge.” Even though the real participants weren’t really hurting anyone, they certainly believed they were. Players who gave louder and longer penalties were clearly behaving more aggressively than those who gave less extreme penalties.
Now let’s take a look at the results for these three different measures of aggression. The different rating systems have been converted into z-scores, which is a way of placing numerically different scores onto a similar scale so that they can be compared. In each case, a greater z-score corresponds to more aggressive behavior or attitude.
It’s clear that the version of Carmageddon 2 that rewarded violence led to the most aggression. Though the difference between the reward/punishment scores was actually not significant in the aggressive affect measure, it was in the other two measures, and in every case, the version that rewarded violence led to more aggression than the nonviolent version. However, it’s important to realize that the negative z-scores in this graph do not suggest that a nonviolent game leads to decreased aggression—this is an artifact of the way z-scores are computed. All we can say from this graph is that playing nonviolent games leads to less aggressive behavior and attitudes than playing violent games.
Many defenders of violent video games argue that they are in control of their own actions, and that video games should not be restricted due to statistical correlations between playing violent games and aggression. In this, they may be correct, but that does not invalidate the results of this study. Playing games that reward violence leads to more aggressive behavior and attitudes than playing games which do not reward violence. Even if this knowledge doesn’t reduce the number of people playing violent games, the knowledge can be used in other ways.
For example, I talked to my son Jim about the study. He’s a typical 13-year-old who plays games that, while not as gory as Carmageddon 2, certainly offer their share of violence. When I pointed out to him that he is more likely to behave aggressively after an especially vicious gaming session, it was difficult for him to disagree. After he figured out that I wasn’t going to take away his favorite games, he even admitted that it was a handy bit of knowledge to have.
Carnagey, N.L., & Anderson, C.A. (2005). The effects of reward and punishment in violent video games on aggressive affect, cognition, and behavior. Psychological Science, 16(11), 882-889