Monday, May 16, 2011

Video Games and Violence: The Debate Rages On

We're combining science and video games! (source)
I think people who play video games are well aware of the public debate over the link between violent video-game playing and actual violent behavior, and if you're not, you can easily catch up with the more than decade-long discourse here, here, and here. What I think most people are not aware of, is that this same debate goes on within the walls of the ivory tower of academia!

I think the debate centers around two key issues: The first is, are violent video-games causing violent behavior? The second is, is the link between violent video-game playing and actual violence strong enough to warrant law and policy changes regulating video-game violence? Let's get right to the evidence after the jump break!

Does violent video-game playing cause actual violence?
Before we get to the evidence about this, let me deal with the issue of correlation versus causation. A lot of lay theorists and armchair psychologists will say, "I knew that [insert name of violent person] was going to snap because he played violent video games such as Halo or Call of Duty," but here's the thing, that is just a single example, and a correlational one at that. All a correlation tells you is that two things go together, not which one causes the other. The only way to know if violent video-game exposure really causes violence, is to randomly assign people to two different video-game contexts of equal arousal level, but with only one of the conditions containing violent action. After participants are placed in two video-game contexts, their aggressive behavior and emotion can be measured to determine if indeed playing a violent video-game increases violent acts.

The good news is that many studies have been conducted already where researchers experimentally manipulate video-game violence-- in the manner in which I just mentioned-- and measure violent acts and emotions subsequently. I won't get into all the gory details of each study. Instead I'll tell you the results from what I view as the very best meta-analytic review of these studies. The meta-analytic review, conducted by Anderson and colleagues, is ideal because it takes the results from all studies (rather than just a single study) examining video games and violent behavior and averages them to give us the best estimate of the true effect of violent video game playing on actual violence using all the available evidence.

The review is unequivocal: there is a clear causal link between violent video-game playing and actual violent behavior, aggressive emotion, physiological outcomes associated with aggression, and aggressive thoughts. Moreover, the review contends that the better the design of the study (e.g., the more tightly controlled the experimental procedures) the stronger the causal link!

To be fair, this conclusion is not without its detractors in the academy. A separate research group, led by Christopher Ferguson, contends that the actual size of the link between violent video-game playing and actual violence is much smaller of an effect than Anderson and colleagues contend. Basically, Ferguson and colleagues argue that the procedures the aforementioned review used caused bias in their meta-analysis.

My personal opinion would be to ignore this criticism for three reasons: (1) Both research groups agree that video-game violence is a causal risk factor for violent behavior. (2) The meta-analysis by Anderson and colleagues represents a larger collection of studies than any other previously conducted review on the same topic. And (3) there is a large history in social psychology research suggesting that exposure to violence increases violent acts, especially among children (Go here to see video from a famous psychological study in which violent behavior towards a toy, modeled by adults, led to violent behavior of children toward the same toy). In essence, the idea that exposure to violence is a causal factor in violent behavior is not a new or controversial idea.

Is the link between video-game violence and actual violence strong enough to warrant policy change?

So, we've now established that violent video games are a causal risk factor for increasing violent behavior. That's fantastic knowledge, but what-- as policy-makers-- should we do about it? This I admit, is a hard question, and one I'm afraid psychological science is not ready to answer yet. The reason is simple: It's one thing to show a causal link between violent video-game exposure and violent intentions, actions, emotions, and physiological profiles, but, it is a whole new issue to determine how exactly specific types of violent acts are linked to video-games AND how to prevent those violent acts. Right now, the science is in the initial stages and I think researchers are just beginning to understand the strength of the causal risks of video-game violence.

The next step for the media and for researchers is to move beyond the simple debate regarding whether video game violence (or violence of any kind for that matter) begets more violence. We already know that based on more than 30 years of research. Instead, new questions should include, for example, (1) understanding how much exposure to violence in media is too much? (2) what the long-term effects of violent video-games include? (3) what are the personality risk factors that make people particularly likely to seek out violent media and respond to it violently? and (4) how young is too young to be playing violent video-games?

 Do you or your children play violent video games, or watch violent media? How do you handle these situations in your home? We'd love to read your comments!

Anderson CA, Shibuya A, Ihori N, Swing EL, Bushman BJ, Sakamoto A, Rothstein HR, & Saleem M (2010). Violent video game effects on aggression, empathy, and prosocial behavior in eastern and western countries: a meta-analytic review. Psychological bulletin, 136 (2), 151-73 PMID: 20192553


  1. Now even economists are getting into the debate. ;-)

    Michael R. Ward, “Video Games and Crime,” Contemporary Economic Policy, 29(2) (April 2011) 261-273

    Finds crime falls with more video game play.

    Michael R. Ward, “Video Games and Adolescent Fighting,” Journal of Law and Economics, 53(3) (August 2010) 611-628

    Finds that the raw correlation is due to "selection bias" and that the conditional correlation is negligible.

    A. Scott Cunningham, Benjamin Engelst├Ątter, and Michael R. Ward, “Understanding the Effects of Violent Video Games on Violent Crime,” April, 2011. <>

    Finds support for an aggression effect of violent video games leading to violent crime, but a larger "incapacitation" effect in which time spent playing video games diminishes crime.

  2. Thanks Michael! I look forward to reading these. Also I love the label "incapacitation effect!" Indeed it appears associations between video game playing and actual crime are more complicated than they are sometimes made to be.

  3. I did a lot of research on this topic, and found an endless spiral of appeals to authority on the part of believers in the media violence hypothesis. The results are not unequivocal (as you put it). Far from it. Try reading: Freedman (2002) and Trend (2007) for an alternative explanation of what you are looking at. Anderson is by far the /best/ media violence researcher, but is guilty of employing ockram's broom in his articles, and shamefully, in his review articles as well. (Okcram's broom is where you brush important counter-evidence under the rug, thereby making your argument appear more convincing.)

  4. Thanks for your comment Aaron. I'd say that the thing about the Anderson et al. review that sets it apart from the books by Freedman (2002) or Trend (2007) is that the Anderson et al. review had to go through a rigorous review process where it was evaluated by anonymous experts in the field. After careful vetting by these experts, Psychological Bulletin (the journal) also published rebuttals by other experts raising some of the issues you raised here.

    To see Anderson et al's responses to these issues (including the broom issue), you can go here:

    Thanks for reading!

  5. Many of our teens today are fund of playing video games not knowing that it brings a big impact concerning their behavior towards things around. A parent should be more responsible in raising their kids and they should embrace programs that focus in the welfare of the kids.

  6. Anderson's meta-analysis mainly looks at work he has done, as well as his former students and co-authors. He only reviews people in his camp. Pretty biased!

  7. Glad you dug it !
    So how do we get people to change their minds and to recognize that kids and youngsters are influenced by all this quite new technologies ? How should we push that conversation in our own little worlds?