Violent video games and public policy (Gentile, Saleem & Anderson, 2007)

This came up in my google search alert (psychology video games) and my first reaction was “how do you make a new scholarly journal?” and “Don’t we have too many to read, especially in my case as VG researcher?” Anyways, Gentile & Anderson along with their grad student Muniba Saleem wrote an interesting review article fit for a class and for the audience.


Policymakers and the public have been concerned about the effects of media violence on children for decades. Scientific psychological research can be an important source of information for policy, as the goal of science is to separate facts from opinions. This article reviews children’s exposure to media violence, describes theories that explain the effects media violence could have, summarizes the research on the effects of media violence exposure, and describes several moderators that can enhance or mitigate those effects. These scientific findings provide useful information for public policy, yet there are many barriers to their use, including misunderstandings of how causality is determined in scientific and public health circles and how large the effects are. Finally, the implications for public policy are discussed, including what has and has not worked in the United States, what other countries and the international community are doing, and where opportunities for new approaches for effective policies may exist.

So I’ll just summarize what they wrote:

Their introduction makes references to school shootings and how the public made knee-jerk reactions for restricting anything related to these shootings, including violent video games. However, at normal times, the public seems ignorant and skeptical of media violence research. Sounds like they’re saying hypocrisy. Well that’s mob rule in this era.

Aggression is defined differently from the public and the scientific community. The public’s definition is quite varied and broad, try asking one person to the next, while scientists maintain a single standard definition so comparisons between studies can be made. IMO, the definition itself is sometimes not good enough (Dodge, Coie & Lynam, 2006) in which researchers have to define different types of aggression (relational aggression).

A definition of media violence: “Media violence refers to media depictions of aggressive and violent behavior directed at characters in the media story. Those characters can be human or nonhuman, cartoonish or visually realistic. Fictional, unrealistic, or animated violence is still considered violence if it meets the above definitions.” (p.17) This is such a good definition that I’ll like to include within wikipedia. An interesting note is that the public tend to view media violence according to content, cartoon violence vs. graphic violence. They don’t see many violent acts in cartoons because their criterion is mainly the graphicness of violence. While studies using the scientific definition have found the opposite, suggesting that humour and/or animation might be attenuating influences of seeing violent acts in people.

They addressed the issues of exposure and the contradicting views of cartoon violence between public and science. Using graphicness solely, (looking at Jack Thompson and the rest) isn’t a good indicator of media violence because it’s like scratching the iceberg and other aspects of violence are relevant and accounted for, such as the intent to harm and how media characters feel or think about harming others. IMO, I like shows and movies that while ultra-violent, respect the audience’s intelligence. Like that movie, Children of Men.

What about Japanese TV? I am such an otaku right now that I could easily switch over to cultural psychology. Contexts seemed to be a factor that differentiates Japanese and American TV violence, not the amount. (Please remember the gauging bias of using graphicness) It’s determined that while being very graphic, the emphasis is on the consequences of such violent acts, usually negative. Any manga/anime consumers would agree to that.

Psychological theories of media effect violence are then explained in brief, I’ll just link them to wikipedia:

social learning theory
cognitive-neoassociation theory:

general aggression model:

Next, they discussed an overall summary of media violence effects, of which I will summarized it further into a few sentences, personal comments and observations. They break down the broad effects of media violence into 5 distinct areas: physiological, emotional, cognitive, attitudinal and behavioural. Lengths: short-term and long-term.

Measures of aggression: Early methods used something similar to Milgram’s experiement of shock with modifications to study aggression. Recently, they’re using the Taylor competitive reaction time task, which involves participants reacting as quickly as possible in order to blast a noise to an opponent. The level of noise is set by the participants themselves, so they set whatever level they want against their opponent. This sounds nice as a way to measure physical aggression. Other measures include teacher/peer/parent/self reports of aggression.

How these different measures correlate? Well, I do hate it when they refer to their own studies or that it can’t be helped since they’re the few ones in psychology studying the relation between violent and aggression . Anyways, they found that they do correlate with each other. I.e. people assessed by their peers/parents/etc as aggressive are aggressive in lab settings. One interesting argument they have is that real life and lab setting aggressive behaviours share a common concept: the intent to harm others.

They recently published a book that focused on three types of empirical research: experimental, cross-sectional and longitudinal. Here, they just give background info and some examples studies.

Experimental: experiments allow researchers to maximize control variables of interest while minimizing or controlling any external influences. However, a weakness is that they can’t measure or induce serious physical harm as it would be unethical, so there are no fights or even pillow fights. Reasons are quite simple, participants have a right to safety, we must minimize chances of permanent harm, and are the benefits outweigh the costs? Another weakness is that experiments look for specific effects, so results from an experimental study is not likely to explain a lot of what happens in real life, just a small bit.

I don’t know why 20 minutes of game play is the standard in VG experimental studies? What is the basis for that anyway? Do kids play at an average of 20 minutes? I don’t, my average is close to an hour per session.

Cross-sectional studies: In general, it’s harder than experiments because researchers can’t control for variables that might influence their results, such as playing history, genetics, temperament, intelligence, media literacy, and social support.

Longitudinal studies: long-term studies are generally expensive and time-consuming. However, the rewards of a good longitudinal study can show the causal inferences of relevant variables using the same participants in their study over a period of time rather than one-time only results of experimental and cross-sectional studies. They presented longitudinal studies that showed the effect of media violence and aggressive behaviours in later life, with some studies saying that trait aggression has no or little role in media consumption. In regards to video game longitudinal studies, there’s only a recent study which found that early violent video game play would lead to more aggression in later life (after 6 months), early aggression was controlled.

An important consideration for aggression research is the cultural context of individual in different countries. While studies from other countries have found similar results of increased aggression from consumptions of violent media. It also touches the concern that most media aggression studies is too U.S. centric and it is something that can answer the question of “TV violence is a reflection of our society’s values and attitudes.” One study has found that children raised in high-aggression/crime/risk/ or warzone areas have higher preferences for aggressive situations and media. (Groebel, 1998)

This is evident on the differences of opinions between American and European research on media research. However, it is noted that some European countries have little research, while others do. But what caught my attention are the differences between sociological research and psychological research. “Sociological approaches tend to relate children’s media use and enjoyment to the social context within which it occurs, whereas psychological approaches tend to focus on the functions media serve and the effects they have. Thus studies that at first appear discrepant are often in actuality complementary. In this context, much of the European research appears more “positive” than the American research.” (p.29) I’d like to make a reference to Sternheimer’s work of which she’s of the sociological circle.

Then they compared the effect sizes of media violence with other medical disorders. This was criticized by Christopher J. Ferguson and Jerald Block (see [1], [2], [3], [4] for bad methodology.

Some things the authors had to say for critics of VG research that they need to consider (in my words):

  1. Studies don’t say that media violence is the single cause of violence
  2. We don’t rely on one type of study to support our theories, we know the pros and cons of different types of studies.
  3. Don’t say we’re biased because you think our studies are crap, while other studies that support your beliefs are good.
  4. How can you criticize VG research if you don’t grasp the fundamentals of psychology?

Moderating factors of media violence effects

  1. Cognitive cues: The more you see violence or aggression, the more you think in aggressive terms. (The more you play WoW, the more you think in WoW terms!) ( see boy and a moose)
  2. Consequences: an aggressive act goes unpunished or rewarded would lead to higher likelihood for aggression.
  3. Justification: my actions are not immoral because that’s how it’s done.
  4. Identification: I’m superman! I’m a tank! I can kick your ass like [insert hero name]!

IMO, not just one or all moderating factor can necessarily justify aggressive behaviours.

What needs to be examined or received little support as moderating factors:

  1. Realism: The more realistic it is, the more aggressive a person becomes (the problem of defining realism.)
  2. Youth vulnerability: kids are more influenced by media violence because they are less experienced or cognitively grown up (studies do point to that effect, but the authors believe the meta-analysis for this idea has some problems and has some doubts about it because of developmental changes in children) or people who can tell from fantasy or reality are immune to media violence (again, people who clearly see fictitious violence are still affected).
  3. Trait aggression: video games only affect those in the high trait aggressiveness range. (that depends on your dependent variable of aggression in your study e.g. physical fights or aggressive thoughts, but they argue that everyone is affected)
  4. Sex: Boys are more affected by violent video games than girls. (but’s that based on earlier studies and they are less aggressive and consume less violent media. But, when they do consume high amounts of violent media, they are more aggressive.)

Violent video games and its moderating factors

  1. Identification with aggressor: Now I agree that players in FPS have to take on the role of the protagonist and have to kill. This would lead to high aggression. Some would wonder why would anyone want to identify with the character, well they don’t. But I say what about the unconscious part? IMO, Now if we can break down character identification without break the FPS experience like cutscenes or something else…
  2. Active participation: Living the experience is a better learning experience, than reading a book. The same is true in video games, especially if you are learning aggressive acts or stuff. IMO, what can we do to learn something other than violence, oh how about learning whether a hammer +2 is better than a sword +2.
  3. Rehearsal of aggressive act: so if I point this gun like this, I get a better result. That means I’ll be a better crack shot at killing people, only if I have a murderous rage and intent.
  4. Reward of violence: shoot him and get 5 points. Most popular FPS games I know of don’t keep a score and most FPS games set the player in as someone with jobs that involve killing people, i.e. military. Those that do keep scores are kept in the same manner as sports.
  5. Repetition increases learning: If you want a perfect game or want to beat insane mode in a game, you gotta learn. Coincidently, you’ll also learn to be more aggressive.

Issues between scientists and the public

Causality and risk factors

If you read the wikipedia articles on causality, probabilistic causation and risk factor, then this is what they’re trying to say. Causality is now a probabilistic phenomenon, not “necessary and sufficient”. Media violence effect weight is based on the weight of other factors as well, that is what risk factor is. “This is not the same as saying that media violence only affects those at highest risk—it means that it is easiest to see and measure the effect on those at highest risk.” (p.41)

Effect size

This is based on methodological disagreement on what is considered a large, medium and small effect size. IMO, there’s the problem of comparing effect sizes from medical science, like the effect size of heart attacks and using aspirin. I’m surprised they did not address criticisms of converting medical effect sizes into psychological science effect sizes. Anyways, they seem to stress the importance of media violence effect as being greater than other health risk factors because most kids are doing games, it is easy to avoid it and it accumulates (but I do wonder if they have that data for video games, they only have one longitudinal study anyway).

Reasons why people don’t seemed concerned

  1. Third-person effect: “These people are affected, but me and my kids are not because I don’t see any changes.” That’s just talk because people aren’t really good at noticing changes within themselves, like assholes. I should also add pluralistic ignorance to the list.
  2. “I watch violent T.V., but you don’t see me around shooting people.” I heard this so many times in’s comments. A good reply would be: “Yes, yes that’s good, but what giving the finger or behaving a lesser form of violence? Are you more of an asshole now than before?
  3. News media giving sensationalist stories and mixed messages that eventually confuse parents into inaction or overaction.

Issues between scientists and the courts

  1. Differences in understanding on some terms like causality.
  2. What is an expert? This allows interests groups to send in people considered “experts” to lobby their case. In the case of the school shooting at Virginia tech, Jack Thompson was called an expert at a news interview, which he gave many wrong facts. So what do you consider a certified expert in the eyes of the court and public?
  3. Courts have different standards than scientists on implementing policy: Does it do immediate harm that justifies impeding free speech? The courts seem less concerned about long-term effects.

Suggestions the authors have

  1. Have these media restrictions on hold until we get more evidence.
  2. More children-oriented shows and more time for kids show, especially on the internet.
  3. Create a public forum to discuss and increase awareness of media violence effect, IMO a rather optimistic and ineffective (?) suggestion. I’d say only the extremists would attend these forums and there’s so much attention we can focus that you’d have to compete with other public concerns, such as cancer, economy or education reform.
  4. Parental monitoring of media consumption. IMO, as if they have the time. If they do, it’s like bringing the kids to a park.
  5. Only spend 1 to 2 hours consuming media. IMO, yeah right you’d have to stop the tv channels and video game producers from making good stuff in order to have my 1 to 2 hours.
  6. Parental discussion of violent TV, that’s sounds good.
  7. More Media literacy programs, that’s what needed for our generation with the internet and all.
  8. “Co-regulation”, combined government regulation and self-regulation (Palzer & Scheuer, 2003)
  9. Create a universal rating system

Here are their reasons:

  1. Current rating system is not reliable or valid (that’s for T.V.)
  2. Parents prefer age based ratings than content-based ratings
  3. But, kids tend to be more interested in older stuff. (i.e. the forbidden fruit effect) Kids and their Halo!
  4. Ratings are based for adults, rather than kids.
  5. Parents have little understanding of the different rating system despite they want ratings systems. (IMO, if parents knew the ratings system a little bit more, then it would preclude the above reasons)

My questions to this universal ratings system:

  1. But you said video games are different from TV because they’re interactive and such, so why create a universal rating if they’re different or is it because they’re not different enough?
  2. Is it because today’s parents can’t handle the different ratings systems?
  3. Is just because of plain old ignorance and carelessness of parents about the rating systems?
  4. Isn’t that a “one size fits all” dilemma? What kind of complexity does this universal rating system would look like?

Well then, I’ll just stop here and rest my head after spending 6 hours and over two days reading and writing about this article…

Gentile, D. A., Saleem, M., & Anderson, C. A. (2007). Public policy and the effects of media violence on children. Social Issues and Policy Review, 1, 15-61.

Dodge, K. A., Coie, J. D., & Lynam, D. (2006). Aggression and antisocial behavior in youth. In N. Eisenberg, W. Damon & R. M. Lerner (Eds.), Handbook of child psychology: Vol. 3, social, emotional, and personality development (6th ed.). (pp. 719-788). Hoboken, NJ, US: John Wiley & Sons Inc.

Feilitzen, C. v., & Carlsson, U. (Eds.). (2003). Promote or protect? : Perspectives on media literacy and media regulations. Göteborg, Sweden: UNESCO International Clearinghouse on Children Youth and Media Nordicom Göteborg University.


3 thoughts on “Violent video games and public policy (Gentile, Saleem & Anderson, 2007)

  1. Pingback: Technological advancement and violence exposure (Ivory & Kalyanaraman, 2007) « VG Researcher - Psychology

  2. Scientific psychological research can be an important source of information for policy, as the goal of science is to separate facts from opinions.

  3. Pingback: Define saleem | Zybis

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s