How consistent are these effects of video games on behavior?

Usually, when I write about research in this blog, I focus on one or two papers that each report a small number of experiments. When psychologists want to know whether a set of results is consistent across studies, thought, they use a technique called meta-analysis. In a good meta-analysis, researchers scour the published literature as well as looking for well-designed unpublished studies that address a common question. The studies are then analyzed as a group. These analyses also use statistical techniques to control for the file-drawer problem. That is, there is a bias for studies to get published if they show a reliable relationship between variables-say between playing violent video games and being aggressive. If a study is conducted and it finds no relationship at all between these variables, that study is less likely to get published. So, studies with no reliable effect are likely to be under-represented in the published literature. Without some statistical correction for the file-drawer problem, then, a meta-analysis might overestimate the true relationship between variables.

Ok, armed with a brief introduction to meta-analysis, let's return to violence and video games.

An extensive meta-analysis of the influence of violent video games on aggressive behavior by Craig Anderson, Brad Bushman, and a number of colleagues was published in the March, 2010 issue of Psychological Bulletin. This meta-analysis examined 380 studies that involved over 130,000 participants. The studies included experiments in which people played video games and then their behavior was measured, correlational studies in which the amount of video game play was related to behaviors, and longitudinal studies in which people were followed over time and changes in behavior were related to amount of video game play. The studies involved participants both from the United States and Japan.

Now for the results.

Looking across studies, playing violent video games does increase aggressive behavior. There are two different mechanisms here. First, playing a violent video game for a short period of time seems to activate or prime the idea of violence. It also increases people's overall level of energy or arousal. So a short amount of time playing a violent video game is generally related to a higher level of aggression immediately afterwards.

Second, longitudinal studies suggest that people who play violent video games for a long period of time are more aggressive overall than those who do not play violent video games. In addition, a small number of studies found that playing violent video games makes people less sensitive to the negative aspects of violence. So, playing violent video games repeatedly seems to teach people strategies for acting aggressively, and at the same time makes them less sensitive to the negative aspects of violence.

There are two other interesting findings from this meta-analysis worth mentioning. First, violent video games seem to have the same effects on men and women. That is, both men and women exposed to violent video games increase their level of aggression. However, it is also the case that men (and boys) are much more likely to play violent video games in general than women (and girls).

Second, there is an effect of culture in these data. Overall, both Americans and Japanese tend to be more aggressive if they play violent video games than if they do not. However, in longitudinal studies that follow people over time, these effects are stronger in American than in Japanese participants. The authors speculate that when violence is depicted in Japan, it tends to be more likely to focus on the negative outcomes of that violence than when it is depicted in America. However, it is clear that a lot more work is required to understand this difference.

Finally, it is not entirely clear what to make of these results. On the one hand, there is clearly an influence of playing violent video games on aggression. The effects are consistent, but they are also relatively small. Obviously, one goal of science is to generate accurate measurements of a phenomenon. This meta-analysis demonstrates clearly that playing violent video games does increase aggression both in the short term and in the long term. But, a second goal of science is to help figure out what should be done given a set of findings. To help answer this question, it will be important to better understand how much playing video games affects the behavior of the people who play them. Answering that question requires more research.