Virtual Violence

Games and aggression have been a hotbutton item for years, and the controversy isn't going away any time soon. Some of it is fueled by conservative paranoia and some by justifiable parental angst. Nevertheless, as Richard pointed out recently, it doesn't mean that the questions are automatically without merit. So I give you this to chew on--a study of gaming and violence that specifically tested an MMOG, did it out of a lab and used a control group:

Hot off the presses as of yesterday in a mainstream, peer-reviewed journal:

Williams, D. & Skoric, M. (2005). Internet Fantasy Violence: A Test of Aggression in an Online Game. Communication Monographs, 22(2), p. 217-233.
Research on violent video games suggests that play leads to aggressive behavior. A longitudinal study of an online violent video game with a control group tested for changes in aggressive cognitions and behaviors. The findings did not support the assertion that a violent game will cause substantial increases in real-world aggression. The findings are presented and discussed, along with their implications for research and policy.

You can download the article from the Taylor & Francis website. The less ethical might look here. I hope that readers here will take the results in context. This is one game and other titles may differ. It's also not a test of young children, but is rather a typical MMOG sample with a broader age range. And (he hints), other research coming out of this same study shows that the game has other effects that are both good and bad. I'll save that for another post.

Personally, I'm still very agnostic about the range of effects that "games" may have. I think that a solid longitudinal test of children probably will show aggression effects (once someone does one), and a host of positives as well. To my mind, they aren't mutually exclusive.


Comments on Virtual Violence:

Edward Castronova says:

Thank God somebody is finally doing rigorous research on this. If I see one more study where violence and media are linked without any attention to causality, I will go bonkers. The causal problems are dual:

1. When violent kids play violent games, it could be the violence in the kid that makes the gameplay especially violent (you don't HAVE to choose the character who has a bloody finishing move).

2. Or, there could be a third factor *cough*homeenvironment*cough* causing both the kid's aggression as well as his preference for violent videogames.

Looks like this study, because it uses a control group methodology, avoids these causality problems. Have to give it a closer read though. But anyways, I am just grateful that someone is finally paying attention to statistical methodology. Hallelujah.

Well. Now we can sit back and watch the newspapers not report this finding.

Posted Jun 14, 2005 4:11:19 PM | link

Flatfingers says:

Dmitri> "conservative paranoia"

Not to be hypersensitive on the subject, but was this label necessary?

For that matter, is it accurate? Are there really no "liberals" who object to violence, whether in video games or Iraq? In the U.S. Senate, I seem to recall Joe Lieberman (not a Republican, last I checked) being one of the folks who lectured video game developers for the content of their products....

That said, although there's no problem with doing a study like this, what does it actually tell us? The concern (as I've said before) from the thoughtful right isn't some simplistic "Rrrrr, this game BAD". It's that these games taken as a group, along with numerous other entertainment media offerings that contain gratuitous violence, contribute to a general coarsening of the culture.

So if this study found no direct link between violent video games and violent behavior in the kids who played those games, that's great and I'm glad to hear it... but it doesn't let game developers off the hook. Just because no individual member of a group is directly responsible for the effects caused by the entire group doesn't mean that individual is free of all responsibility for their actions.

And no, I don't expect this to be a popular position to take on this issue. But it's a serious one.

--Flatfingers

Posted Jun 14, 2005 7:12:35 PM | link

Cory Ondrejka says:

Wow, Dmitri . . . very impressive! Sadly, as Ted already noted, certain to be non-news.

Finally someone doing a study that goes beyond the decades-old work of Aronson et al and actually looks at the impact of exposure. Such a refreshing change from the usual presumption that the short term increases in aggressive behavior long noted to follow exposure to violent media will lead to increases in violent behavior.

Bravo -- and I can't wait for more research!

Posted Jun 15, 2005 12:32:28 AM | link

Matt Mihaly says:

Flatfingers wrote:

In the U.S. Senate, I seem to recall Joe Lieberman (not a Republican, last I checked) being one of the folks who lectured video game developers for the content of their products....

Being conservative and not being Republican are hardly incompatible properties. Joe Lieberman is one of the scariest (in my opinion) conservatives out there.
--matt

Posted Jun 15, 2005 1:14:59 AM | link

Theo says:

Well said Edward. I'm always very skeptical of studies and statistics for precisely the reason you mentioned: lack of causality. If you can't learn anything meaningful from the statistic, you're almost better off not to have it in the first place, lest you find yourself mislead.

Posted Jun 15, 2005 8:39:45 AM | link

Arnold Hendrick says:

I found the study very interesting, especially the discovery of very slight tendancy for game violence to have a slightly greater effect on OLDER players. The authors opined that older "Baby Bommers" might not have been exposed to as much violent game fare as younger generations, and therefore were not innoculatd to its effects.

A more interesting issue in my mind is the potential "impersonalization" of other players in heavily PK-enabled worlds. For example, in a game like Lineage II (which is fully PK-enabled) some players avoided PKing because they knew it would cause discouragement and upset to the players they victimized. However, others clearly enjoyed ganking weaker, virtually defenseless opponents. While I suspect the game simply enables hostile, selfish behavior already present in certain people, would exposure to such acts cause one to seek "revenge" and become more violent too?

An even more complex analysis is possible in games where people can select or avoid PK environments, such as WoW or DAoC.

Posted Jun 15, 2005 4:09:51 PM | link

Thabor says:

I'm grateful that someone dedicated time to doing real science, instead of generating headlines.

Posted Jun 15, 2005 4:24:57 PM | link

Flatfingers says:

Here's a link I noticed recently to another study on computer games and aggression: Do Games Prime Brain for Violence? in New Scientist News.

Note: This isn't a hit piece; it's just another research citation for those interested in this subject. The writer of the NS article actually takes a fairly neutral position on the question.

--Flatfingers

Posted Jun 27, 2005 11:27:50 AM | link