Video games, youth violence and crime

This study “A longitudinal study of the association between violent video game play and aggression among adolescents.” by Willoughby, Adachi, & Good, (Dev Psychol. 2012 Jul;48(4):1044-57) was today hot topic at Facebook. I am skeptical about simplistic claims (causally) linking violent video games to aggression and violent behavior and did quick search on other related statics. I found today a page http://videogames.procon.org/view.resource.php?resourceID=003627.

The data in that site seems to indicate that there are no link between game sales (I assume that violent video game sales follows same trend that video game sales) and violen crimes, youth violence, and bullying (see figures 13–17).

But what is worrying in these figures are in comparing problem behavior between the players of M-rated games to those who do not play M-rated games. The M-rated game players have statistically higher rates of problem behavior in various areas such as being in fight and stealing (see table 21). However, the amount of students reported to being in fight has not changed much (see figure 20).

I am not sure what make out of this. But here are couple toughs:

  • Kids who play M-rated games are prone to take more risks than those who do not play.
  • Violent games are played more by those that are prone to aggressive behavior.

Unethical Research?

Malena Ivarsson et al. put 12-15 year old children to play Animaniacs and Manhunt to see if there is different effects on playing these games (http://www.su.se/forskning/ledandeforskning/samhallsvetenskap/stress-arbete-och-halsa/valdsamma-dataspel-paverkar-kroppen-1.79509)

Can you see any issues here?

The Pegi rating of the Manhunt is 18, because it contains extreme violence. Hence, it is not meant to be played by 12-15 year old children. Why it has been judged to use this game and 12-15 year old children necessary?

A  publication by Ivarsson et al. (2008), “Playing a violent television game affects heart rate variability” (in Acta Pædiatrica) assumes that the games are comparable, because they did not find differences between the games with an Actiwatch in motor-patternsof playing.Yet, they failed to account the qualitative differences between games, as Annika Waern point out on her blog. (I assume that it is a part of the research promoted in the above-mentioned su.se page as the measurements, games, and main authors are the same that are mentioned on valdsamma-dataspel-paverkar-kroppen-1.79509)

The study does not do anything to control the qualitative factors. It is possible that the better or just more intensive game have more lasting impact despite the content, isn’t it?

Waern continues:

[T]heir second study is more interesting than their first. Remember that the second study showed that the boys who played little games reacted stronger to Manhunt than those who played a lot of games. Ivarsson interprets this as a desensitization towards violence (in general), but there are at least two other possible explanations. One is that the boys who played a lot of games understood the horror genre better, thereby getting less scared. But the other explanation, and the one I think is the right one, is that they were better at understanding the game as a game.

These are possible alternative explanations to desensitization. In addition, just be able to prepare oneself for the future events help.

Ivarsson is rather explicit on her presupposition:

Malena Ivarsson am personally convinced that violence in computer games in the long run affects the players, and the subtle but gradually can contribute to changing behaviors. (cited in Lundström, translation by google)

I have iterated before why I think that the above premise about a simple causal connection where violent videogames are claimed to cause violent behavior is not plausible.

In the end, based on Ivarsson et al (2008) we can conclude that these two game players had different heart rate patterns and possibility some sleeping issues. So, the authors show evidence to support their premise (in this specific case). But what that means in general and how, for example, this study supports claims on valdsamma-dataspel-paverkar-kroppen-1.79509 or what Waern summarizes on her blog?

There is a big leap from the results to claims presented on valdsamma-dataspel-paverkar-kroppen-1.79509.

Violent Video Games Effects? What Are They?

Now game violence effect discussion is active again in Sweden after Karolinska Institute researchers have been publishing their opinions in DN, I did some research on the topic (again).

A report Understanding the Effects of Violent Video Games on Violent Crime by Cunningham et al (2011) states:

First, they [the study results] support the behavioral effects as in the psychological studies. Second, they suggest a larger voluntary incapacitation effect in which playing either violent or non-violent games decrease crimes. Overall, violent video games lead to decreases in violent crime. (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1804959)

I previously posted about a review study that links personality traits and video game violence influence, but the authors posists following reservation:

Given the number of youths who regularly engage in VVG play and the general concern regarding this media, it would seem likely that resulting violent episodes would be a regular occurrence. And yet, daily reports of mass violence are not reported. It appears that the vast majority of individuals exposed to VVGs do not become violent in the “real world.

US Supreme court takes very similar stance that the two above studies:

Psychological studies purporting to show a connection between exposure to violent video games and harmful effects on children do not prove that such exposure causes minors to act aggressively. Any demonstrated effects are both small and indistinguishable from effects produced by other media.(http://www.supremecourt.gov/opinions/10pdf/08-1448.pdf)

Games Briefs has an interesting figure on the topic (video game sales vs violent crimes / US figures). These figures are also in the line what said in above.

While there is definitely need for research in the area (as well as restriction for selling games for minors), it should be obvious that the effects of playing are not very straightforward. If they are, we would be seeing rabidly raising amount of violent crimes even with small effect sizes.

Gerald Jones have a good critical take on subject in his book Killing Monsters! Why Children Need Fantasy, Super Heroes, and Make-Believe Violence were he argues that make-believe violence has a role in children development.

EDIT:

There is also methodological issues in effect studies conducted to children (from the Gerald Jones’s book; I do not have book now, so no pages numbers.).

  1. The test setup can be cause of aggression itself
  2. Make-believe (playing about what was just seen) is interpret as aggression.

EDIT2:

Gerald Jones discusses the issue on media effect studies on pages 23-44 (Killing Mosters).

EDIT3

Links to the discussion in DN (in Swedish, not exhaustive)