Callusing blog header photo
Callusing's c-blog
Posts 0Blogs 7Following 0Followers 2



How to argue about video game violence: A primer

The gaming community has a violence problem. The video game industry has grown into a behemoth, boasting $67 billion a year in income, counting 80% of youths and adolescents among its clients, and spending millions each year on lobbying. Violent games are its bread and butter: between 2008 and 2011, Mature-rated games, most of them rated so due to violent content, rose from 16% to 27% of all game sales, and five of the top ten best-selling games of 2012 focus on digital murder. As the number of massacres marches upward and our public figures look for ways to combat violence and aggressiveness within American youth, the finger has been pointed squarely at violent video games.

The gaming community's response has been as intense and impassioned as it has been juvenile, with a level of conversational depth that sometimes lies between a middle school food fight and an Infowars forum thread. Despite the apparent intelligence and dedication of most community members, their arguments often scrape the bottom of the rhetorical barrel, relying on strawman arguments, burden-of-proof claims, and ad hominem attacks aplenty. To say these arguments have been ineffective is an understatement; while the debate has never been entirely civil, the gaming community's attacks on politicians and dismissals of widely-held beliefs have served to only reinforce the preconceived notions of those who wish to paint gamers as reckless media addicts. Both science and policy have ground to a standstill, with academics on both sides openly voicing their frustrations. 

This is not sustainable. The gaming community is, at best, throwing away an opportunity to have its voice heard and at worst is degrading the chances of escaping without restrictive legislation. This conversation can, and must, change. The level of discourse needs to be elevated, and the gaming community needs to transform its image in this debate from a group of teenagers and young adults with a bone to pick to a community of bright, analytic minds who have something to bring to the table. This elevation depends on a shift in thinking away from reflexive responses, denials, and name-calling and toward progressive suggestions and beneficial analyses.

This is not a summary of the literature. Professional academics have already authored several in-depth meta-analyses and reviews, and there is little I can add to that discussion at this time. This is instead an analysis of the rhetoric of the debate, focusing on the arguments used by the gamers and supporters of violent video games. I hope to define why particular lines of argument are and are not valid, and to provide a structure for future debate and research. And while I personally believe that the literature is fairly equivocal on whether games cause violence and aggression, I hope that the way I lay these arguments out will be clear and understandable to people who lie on any side of this debate.

Equip yourself with these tools, and you'll join the company of the other community members who can make a difference.

Arguments that work

Observational studies can't demonstrate cause and effect

Example The fact that the 300 subjects that played the most video games had a higher risk of committing violent or aggressive acts does not show that video games cause aggression.

Psychological research falls into two major categories--observational, which is further subdivided into cross-sectional and longitudinal studies, and experimental. Cross-sectional studies compare two or more groups at a single point in time: “Group A played 20% more violent video games, and committed 10% more acts of violence than Group B.” Longitudinal studies track a single population over time: “Across the group as a whole, an increase in violent video game use tended to precede an increase in violent acts.”

No matter how rigorously they are performed, these studies cannot demonstrate causation, only correlation. This is best illustrated with an example. Imagine a study where 300 subjects of varied backgrounds are surveyed on their use of violent video games and are tracked for incidents of violent behavior for 1 year. Imagine, furthermore, that this study shows that the subjects who played video games for more than 2 hours a day were 20% more likely to commit an act of violence than their non-gaming peers. While it is tempting to say that the video games caused the violence, we can't actually draw this conclusion; instead, we can only say that the kind of person in this sample who plays video games for more than 2 hours a day is also the kind of person who is more likely to commit a violent act. In other words, we can't tell whether the video games caused the violence, whether a violent nature drew the participants toward violent video games, or whether some third variable (such as the amount of free time) caused both effects. A longitudinal study comes closer to causation, but doesn't bridge the gap; even if a study shows that video game use consistently precedes an increase in violence, it's impossible to tell whether the video game use is causing the violence, or whether some unknown variable (a confounder in technical terms) is causing both video game use and violence. 

This goes beyond a purely academic concern. A number of observational studies have concluded that violent video game use moderately increases violence and aggression in youths, lending support for anti-game legislation. If video game use is not causal, however, legislating against it will not only be needlessly restrictive, but will also be ineffective at reducing youth violence. While these observational studies can generate the hypothesis that violent video games cause violent tendencies, rigorous experimental studies are needed to show causation.

As a brief digression, some researchers have pointed out that the fact that smoking causes lung cancer, despite being widely accepted, has only ever been studied observationally. The connection between smoking and lung cancer, however, is quite different from the connection between video games and violence. Lung cancer is all but absent in the non-smoking population, but many violent individuals have never touched a video game; furthermore, observational studies have consistently shown that smoking increases lung cancer risk by 20x or more, while video games, if they have any effect at all, have a much less significant effect on violence. So while we gave smoking judgment a free pass without experimental research, the same cannot be done with video games and violence. We need experimental studies.

Indirect experimental studies may be inapplicable to real-world effects

Example Those survey results tell us that the people who played violent games seem more aggressive in a clinical setting, but we don't know how well this research applies in the real world.

An ideal experimental study randomly divides subjects into two (or more) groups. Randomization eliminates the problem observational studies have with confounders; if the groups are large enough, researchers can assume that confounders will be evenly distributed between groups and their effects will cancel out. One group is given a treatment, while another is not. If the study is well-controlled (meaning the introduction of new confounders is prevented), any differences in outcome between the groups can be described as caused by the treatment. For example, if researchers took 600 subjects and told a randomly selected 300 to play violent video games for 1 hour per day, and the 300 remaining to not play any violent games, a strong case could be made that any increases in violence or aggression seen in the former group were caused by the video games.

There's an issue with studying violence directly--violent acts are incredibly rare. In 2011, there were only 423 arrests for youth violence for every 100,000 youths. This means that, even if a study tracked 2,000 adolescents for a year and violent video games (as a treatment for one group) doubled the rate of violent actions, the treatment group, on average, would only have 8 events, compared to 4 in the control. To see a difference from treatment that goes beyond what could be caused by chance, a study would need to have an enormous number of participants--which would result in an enormously high cost (in the many millions of dollars).

So instead, researchers attempt to use proxies for violence, including various measures of aggression. These proxies range from mundane, like surveys that ask participants how they feel about certain actions or ideas, to eclectic, like using a subject's tendency to dole out painful noise blasts or hot sauce to measure their aggressive nature. Whether the study is over a single playing session, a few days, or a longer period, researchers use these measures to determine how aggressive or violent an individual is without forcing them to actually act aggressive or violent.

There is a tremendous amount of debate surrounding the applicability of these measures. While some argue that a short-term demonstration of aggressive tendencies is relevant to the real world, others argue that these lab measures, whether brain scans, surveys, or spice rationing, tell us little about the way these subjects would actually act. And while the value of these markers could be tremendous, the burden of proof falls squarely on those who use them. Until it has been demonstrated that these markers truly represent real-world behavior, these indirect experimental studies will remain in question.

Research has been looking at the wrong people

Example If we think that video games may only affect a subset of the population, are population-wide studies useful?

As mentioned above, violence is rare, and extreme acts of violence like school shootings are even rarer. The latter is so rare, in fact, and the profiles of the perpetrators so varied, that researchers are currently unable to produce an effective profile of the environment or actions of a “school shooter”. Violent video games, on the other hand, are very common, and most adolescents play them without becoming violent. It has been suggested, then, that violent video games may have a heterogeneous effect on the population; in other words, that some subset of the population, such as those already at risk of violent behavior due to other environmental factors, are affected more profoundly by violent games than the rest of the population. If this is true, it has problematic implications for population-wide studies, which will average the effects of a few to a population as a whole, and thus understate the impact of violent games in the highly-affected subset while overstating their impact in everybody else. 

Arguments that sometimes work

(x) doesn't play games, or doesn't know what he's talking about

Example (bad) Jack Thompson has never played violent video games, so how can he discuss them intelligently and who is he to be driving policy? He's just a biased outsider.

Example (better) This article mischaracterizes the way Grand Theft Auto rewards violent actions, so any arguments that depend on this point are not valid.

The majority of politicians who support anti-violent-game legislation are far outside the games' target demographic. It's easy to suggest that these old men don't know what they're talking about, and they make it even easier with statements about “scoring points for hitting people in Grand Theft Auto.” It's worth keeping in mind, however, that the same kinds of accusations are often used by the other side. Chris Ferguson, a social psychologist known for his stance against a connection between violent games and aggression, has been called out for the fact that playing games is "an important part of [his] identity.” Many pro-legislation pundits dismiss arguments by gaming community members on the grounds that they have a vested interest in maintaining their hobby.

But attacking the person does nothing to show why the argument itself is good or bad. While an individual with limited gaming experience is unlikely to know the intricacies of the medium as well as a veteran player, this doesn't preclude them from making valuable arguments. If the important parts of an argument, however, hinge on a misconception, then these misconceptions deserve to be pointed out, and the argument evaluated to see if it still stands without them. For example, the above article, as well as testimony in a notable trial, suggested that players get points (meaning they are constantly rewarded) for killing innocents in Grand Theft Auto IV. Anybody who has played the game knows that not only is this technically inaccurate, but that this entire kind of reward structure is absent from the game (no reward is given for killing civilians). As a result, any arguments or statements that depended on the existence of this reward system are invalid. 

Attacking the person saying the statement blackens the entire argument--both its sound and unsound points--and does nothing to progress the understanding of either party. But targeting the specific parts of the argument that are faulty allows progress to be made.

(x) study or (y) pundit was funded by (z)

Example A study on the gaming brain was funded by a pro-”family values” group, so we should treat it with suspicion

An accusation of funding-induced bias is fundamentally the same as the point above--a suggestion that the argument in question doesn't work because its source can't be taken seriously. But funding, like a lack of subject knowledge, does not doom the study to bias, or as the authors at RPS wisely point out,

...their funding the research does not necessarily bias the research, there’s no evidence to suggest that the research group are anything other than science-focused and pursuing the truth

Dismissing an argument or study because of its funding does nothing to show why it is invalid. Instead, a suspicious study should be inspected for errors. Authors may use methods that bias the study toward particular outcomes; more often, good results are selectively interpreted to generate conclusions that may not truly follow from the data. A close reading of the source material, in other words, can often determine if a biased funding source has led to a biased study.

Arguments that don't work

I'm a gamer, and I'm not violent, so games don't cause violence

Example I played Splatterhouse for 10 hours this weekend, and I feel fine.

Every reasonable person on both sides of this debate would agree that video game violence does not cause every single player to become violent. And all would agree that violence is rare and video game use is common. It's apparent, then, that violent video games, if they have an effect at all, have one that builds slowly over time, has a small magnitude, or affects only a subset of the population. No individual anecdote, whether a nonviolent clan of gamers who have played Counter Strike for 10 years or a pair of teens who played a lot of Doom before taking the lives of many others, is of any significance at all in helping determine which of these four cases is true. It's impossible to know if any case follows the rule or is somehow an exception; it's for this exact reason that good research uses large and diverse samples.

We had violence before we had video games / crime rates have gone down since games came around

Example Over the past decade, violent game sales have increased, but youth violence has decreased, so games can't be causing violence.

These comments are designed to show that video games can't be a significant contributor to violence, since violent acts have not increased over the time frame we've been exposed to video games. It's obvious, however, that the causes of violence are many and their interactions complex. We don't know how much violent video games contribute to current violence rates; in other words, we don't know if a world without violent video games would have more violence, less violence, or approximately the same amount. And even if video games are found to not increase violence on a population-wide scale, they could still feasibly increase the tendency toward violence in a subset of the population. This kind of argument does nothing except demonstrate its user's simplistic understanding of the causes of violence and aggression.

We already know the answers, so why do we need more information?

Example It's obvious to all of us that video games don't cause violence, so why are we wasting government funding on more research?

When the Obama administration agreed to write $10 million in checks to researchers to study the effects of media on violence, many within the gaming community (and some conservative pundits) cried afoul. The debate is settled, they cried, and more research money is a waste. Even a quick survey of the literature, however, demonstrates prominent and respectable academics on both sides of the argument, all of whom agree that we lack definitive answers. Until we gain an understanding of the way violent video games affect behavior, it will be impossible to make wise policy decisions. The solution, however, is not just more research, but better research. Instead of kicking the can down the road with dozens of small studies, we need powerful, definitive studies to answer these questions once and for all. This $10 million grant could give scientists the ability to do this kind of research, and that potential should be tantalizing for everybody on both sides of this debate.

So, what do we do?

Some political theorists have reasoned that war is impossible without the dehumanization of the enemy. Much of the purpose of propaganda draws from this point--by painting the enemy as something less than your nation, it becomes much easier to ignore their viewpoint entirely and follow your own nation with unquestioning allegiance. This debate over video game violence is not a war. It is not “us versus them”. The goal should not be to convince those who want to legislate against violent games that the gaming community is right, but to find the arguments and information that will bring them to the community's side.

By painting itself as a constructive contributor to the conversation, the gaming community can show that it deserves the ear of policymaker and scientist alike. By demonstrating that its members are articulate and reasonable enough to assess and digest arguments accurately, the community can gain a say in a debate currently dominated by academics, media spokesperson, and politicians.

The gaming community can show that it deserves a seat at the table, or it can continue to rely on the same hackneyed arguments and throw barbs from afar. It doesn't take a genius to recognize which of these will make a greater contribution to the discussion.
Login to vote this up!


Sotanaht   1
taterchimp   1
Elsa   1
qlum   1
Ben Davis   1
BrowneyeWinkin   1
JoyfulSanity   1



Please login (or) make a quick account (free)
to view and post comments.

 Login with Twitter

 Login with Dtoid

Three day old threads are only visible to verified humans - this helps our small community management team stay on top of spam

Sorry for the extra step!


About Callusingone of us since 11:20 AM on 06.07.2013

Recent grad, professional researcher, aspiring writer.