Each of these approaches has its flaws. The first kind of correlational study can never prove that video-game playing causes physical aggression. Maybe aggressive people are simply more apt to play violent games in the first place. Meanwhile, the randomized trials, like Anderson and Dill's, which do imply causation, necessarily depend on lab-based measures of aggression, such as whether subjects blast each other with noise. This is a respected measure, but obviously not the same as seeing whether real people hit or shoot each other. The longitudinal work, like this year's elementary-school study, is a useful middle ground: It shows that across the board, playing more-violent video games predicts higher levels of verbal and physical aggression later on. It doesn't matter why the kids started playing violent games or whether they were already more aggressive than their peers; the point is that a year of game-playing likely contributes to making them more aggressive than they were when they started. If we had only one of the three kinds of studies, the findings wouldn't mean much. But taken together, the body of research suggests a real connection.
The connection between violent games and real violence is also fairly intuitive. In playing the games, kids are likely to become desensitized to gory images, which could make them less disturbing and perhaps easier to deal with in real life. The games may also encourage kids (and adults) to rehearse aggressive solutions to conflict, meaning that these thought processes may become more available to them when real-life conflicts arise, Anderson says. Video games also offer immediate feedback and constant small rewards—in the form of points, or access to new levels or weapons. And they tend to tailor tasks to a player's skill level, starting easy and getting harder. That makes them "phenomenal teachers," says Anderson, though "what they teach very much depends on content."
Critics counter that some kids may "use games to vent anger or distract themselves from problems," as psychiatry professor Cheryl Olson writes. This can be "functional" rather than unhealthy, depending on the kid's mental state and the extent of his game playing. But other studies suggest that venting anger doesn't reduce later aggressive behavior, so this thesis doesn't have the most solid support.
When video games aren't about violence, their capacity to teach can be a good thing. For patients suffering from arachnophobia, fear of flying, or post-traumatic stress disorder, therapists are beginning to use virtual realities as a desensitization tool. And despite the rap that they're a waste of time, video games may also teach visual attention and spatial skills. (Recently, a study showed that having played three or more hours of video games a week was a better predictor of a laparoscopic surgeon's skills than his or her level of surgical training.) The games also work for conveying information to kids that they will remember. Video games that teach diabetic kids how to take better care of themselves, for instance, were shown to decrease their diabetes-related urgent and emergency visits by 77 percent after six months.
Given all of this, it makes sense to be specific about which games may be linked to harmful effects and which to neutral or good ones. Better research is also needed to understand whether some kids are more vulnerable to video-game violence, and how exposure interacts with other risk factors for aggression like poverty, psychological disorders, and a history of abuse. Meanwhile, how about a game in which kids, shrinks, and late-night comics size up all these factors and help save the world?