Violence over competition in violent video games (Anderson & Carnagey, 2009)

One of the labs I’m volunteering in has computers with two screens. I never got the chance in trying them out, but it would’ve been nice in reviewing studies: the journal article on one screen and my reading notes on the other; or flipping through blog entries, news, and emails.

Craig Anderson and Nicholas Carnagey of Iowa State University have published a violent video games effects study. The study’s focus is on a counter-argument used by many gamers: competition is also responsible for the observed aggression in video game studies. Their study revealed much of the role competition has in violent video games, however the study’s implications are limited by its methodology.


Three experiments examined the impact of excessive violence in sport video games on aggression-related variables. Participants played either a nonviolent simulation-based sports video game (baseball or football) or a matched excessively violent sports video game. Participants then completed measures assessing aggressive cognitions (Experiment 1), aggressive affect and attitudes towards violence in sports (Experiment 2), or aggressive behavior (Experiment 3). Playing an excessively violent sports video game increased aggressive affect, aggressive cognition, aggressive behavior, and attitudes towards violence in sports. Because all games were competitive, these findings indicate that violent content uniquely leads to increases in several aggression-related variables, as predicted by the General Aggression Model and Inari-zushirelated social–cognitive models.

I suddenly have the urge to eat inarizushi.

I’ll skip the General Aggression Model which explains the mechanisms of all things aggression and go over to competition theory, which is not elaborated much (IMO). Violent video games are generally competitive in contrast to the non-competitive nature of non-violent games. This competitiveness can confound research results and is every so often pointed out by gamers. True, competition has a role in aggression, but it has never been controlled in experiments before. Therefore, Anderson and Carnagey decided to test this competition-only theory by manipulating the level of violence in sports video games.


They conducted three experiments, but they are not at all different from each other so it’s not a comprehensive study of the competition vs. violence problem. Hence, I condensed parts of the study for conciseness.

Participants: 377 participants, about equal gender ratio, I suspect from psych classes since they are given course credit for participation. No other demographics given (age, ethnicity, etc.)


Across all experiments

Video game violence exposure questionnaire: Participants listed five most played video games from 7th grade to present. They rated, on 7-point scale, these games based on violent content. They were also asked how often (7-point scale) they played the following games: Madden Football, NHL 2004, MVP baseball, FIFA Soccer, and ESPN NBA Basketball (nonviolent games); NFL Blitz, NHL Hitz, MLB Slug Fest, Red Card Soccer and NBA Hoopz (violent games).

Sports experience questionnaire: 7-point scale questionnaire on how often they watched and played sports (football, hockey, baseball, soccer and basketball).

Aggression Questionnaire: the standard Buss and Perry Aggression questionnaire. Although, it seems only the physical aggression subscale (9-items) is considered. I wonder why they haven’t used the entire aggression questionnaire in the study.

Video game ratings scale: Participants rate the game they played in the experiment in terms of difficulty, enjoyment, frustration, excitement, pace and violence on a 7-point scale. They also rate themselves on how they performed in the game and how much they improved their abilities from start to end.

Heart rate measure: They measured heart rate at several points in time during the experiment.

Experiment 1-only measure: Aggressive cognition from the Word Pronunciation Task. The WPT is basically the oral version of the Word Completion Task where participants name the word they see on a computer screen. So they would pronounce neutral, aggressive, control words in a random fashion. The reaction time in pronouncing the words is used to determine how quick and accessible a certain cognitive schema is. So if you’re aggressively primed, then you’re quick to say aggression-related words than science-fiction-related words.

Experiment 2-only measures: Aggressive affect (feelings) from the state hostility scale. Participants rated their current mood based on 35 adjectives on a 7-point scale.

Attitudes towards violence in sports questionnaire: participants rate (7-point scale) the appropriateness of various aggressive behaviours in specific sports (football, hockey, baseball, soccer, basketball). It’s a good idea to measure individuals’ attitudes on sports violence since the level of acceptance varies between sports, notably fights occur often in hockey and penalties are light compared to others.

Experiment 3-only measures: Aggressive behaviour from the Competitive Reaction Time Task. Participants must quickly press a mouse button after an auditory cue to blast noise of their own setting to another supposed participant. Participants go through 25 noise blasting trials. The trials were predetermined in that participants would always get 13 wins and 12 losses in a random fashion. Professor Ferguson has written an article questioning its validity. I haven’t read it.

The Competitive Reaction Time Task Motivation questionnaire: This questionnaire measures participants’ aggression motives: instrumental aggression and reactive (“revenge” as it’s written). 6-items.

Video games used: They conducted a pilot study to pick which video games to use in their experiments. They selected nonviolent video games and violent game versions of basketball, hockey, soccer, baseball and football. They found the baseball and football video game suitable for the experiments. Hockey was excluded because both versions were aggressive, both soccer’s and basketball’s violent versions were not “excessively” aggressive enough.

In the end, the violent video games used in the experiment are MLB Slugfest Baseball and NLF Blitz Football. The nonviolent video games used are MVP Baseball 2004 and Madden Football.


This procedure applies to all three experiments. Participants completed the questionnaires, except the video game ratings scale, which comes at the end of the experiment. They are randomly assigned to play one of the video games for 20 minutes. Then, they complete the experiment-specific measures. After, the complete video game ratings scale and then debriefed for suspicion and debriefing. Any participants who suspected would be noted in the analysis and excluded if necessary.

The reason why all the experiment-only measures could not be combined in a single experimental procedure is that participants might catch on to the purpose of the study, which can influence their responses to the questionnaires and possibly lengthening the experiment too long for the participants.


All experimental results are combined.

Video game ratings scale: Violent video games were rated more violent (M = 3.99) than the non-violent ones (M = 2.07), so that’s good. They were also rated as more difficult (M = 4.30 vs. M = 3.94), frustrating (M = 3.99 vs. M = 3.54), and faster pace (M = 4.40 vs. M = 3.75). There were no differences in enjoyment and excitement (please refer to Przybylski).

The authors did not elaborate the differences in difficulty and frustration. I’ll elaborate in their place. IMO, the nonviolent sports game are realistic simulation of the real life sports, so the rules and mechanics are faithfully followed. Game balance is achieved. Changing the sports game to a more violent version would inevitably entail a modification of the game mechanics in order to incorporate the violent content; this means the rules are more relaxed. Such distortions would unbalance the games and confuse players familiar with the sport, making the games more difficult and frustrating.

Heart rate: Nothing of significance found.

Aggressive cognitions: Main effects found. Participants in the violent game group scored higher in aggression (M = 26.28) than the nonviolent group (M = 16.85). Men (M = 26.40) scored higher than women (M = 16.73).

Aggressive affect: They’ve broken down the State Hostility Scale into several subscales.

Aggravation subscale: those in the violent group felt more aggravated (M = 2.40) than the nonviolent group (M = 1.96). In addition, those who rated high the games on frustration and difficult also has a positive effect on the aggravation subscale. Past video game violence exposure was found to moderate the relationship, so those with plenty of violence exposure had higher aggravation scores, but this does not interact with the violence level of the experimental games.

Feeling mean subscale: A statistical trend (not significant enough) was found in that the violent group may or may not have felt more mean than the nonviolent group. Again, game frustration was a significant factor.

Attitudes towards violence in sports: There were no general statistical significances in changes of attitudes towards violence in sports. What they found is that participants in the violent group were more supportive of aggressive behaviours in hockey and men were more supportive of aggressive behaviours in hockey and soccer.

The authors explained that because violence has an ambiguous acceptance compared to others sports, people would not feel strongly about the level of aggression in hockey than other sports. Of course, how well you know the sport also determined the level of acceptance of aggression in a specific sport. The authors explained why participants were more supportive of more aggression in soccer is probably due to lack of popularity in the area of study. Football is quite popular in the U.S.

Aggressive Behaviours: The authors separated the results into two. High intensity aggression and average intensity aggression. Reasons being that high intensity is a clear sign of aggression, it is most likely to instil retaliation and easier to communicate to non-expert audiences. Yes, it’s for you gamers.

High intensity aggression: Noise blast level 8-10 is considered high. There’s significance in that the violent group behaved more aggressively than the nonviolent group. In their words, the violent group gave 75% more high intensity blasts than the nonviolent group or in raw number terms, a mean of 4.65 vs. a mean of 2.65. So that means, if given more trials in the noise blast, their victims would have ringing ears and a headache than those in the nonviolent group. Of course, this is up to interpretation. Anyone?

Men were also more aggressive (M = 4.48) than women (M = 2.82).

Average intensity aggression: On average, the violent group gave higher noise blasts (M = 5.15) than the nonviolent group (M = 4.62). Men were also more aggressive than women. Participants’ rated ability in the game (i.e. performance) is positively related to aggressive behaviour, participants’ difficulty rating is also positively related to aggressive behaviour.

Aggressive behaviour motivation: Reactive aggression (i.e. retaliation) was a significant predictor for high intensity and average aggressive behaviour.


The study experimentally showed that violence has an effect on aggression despite the equal level of competitiveness in both video games. Although the authors noted study limitations, such as they did not examined competition alone (separated from violent content). Therefore, this study does not rule out competition from the aggression effects equation. I think THAT can be arranged with my Zsnes and my collection of roms. I know one very competitive video game that should be of interest, one different from the authors’ idea of competitive gaming.

The authors’ selecting sports video games is certainly a logical choice since the objective of these games is to score as many points as possible while preventing the opponent from scoring points. These games certainly have competitive elements and there are violent versions of these games. This allows keeping (nearly) competitiveness under experimental control and manipulate the level of violence. Unfortunately (IMO), sports video game may fail to capture, by gamers’ standards, competitiveness.

The concept of competition between gamers and researchers are possibly different. Many gamers talk about competition from First-Person Shooter games and Real-Time Strategy games, these genres are often the main focus in live video game competition events. I rarely hear sports video games being treated seriously as FPS and RTS. Furthermore, differences between genres in game mechanics may certainly have a role in competition and aggression. For example, FPS has players fight each other as individual avatars, RTS has players complete control over their units and resourcing. Sports video games have A.I. control assistance as a means to run the game as smoothly and realistically as possible. This kind of assistance might have players feel less competitive since they exerted less control and effort in winning matches. Another criticism is the lack of social presence or the absence of human players. Players would feel less excited and satisfied playing against a computer than a human player since computers are predictable. Matthew Eastin has published a study which tentatively suggested that competition in violent video games may have a role in aggression priming. Therefore, this study’s generalization is limited. Combined with other studies, it certainly shows where video game competition effects weakened or strengthened regarding aggression.

Anderson, C. A., & Carnagey, N. L. (2009) Causal effects of violent sports video games on aggression: Is it competitiveness or violent content? Journal of Experimental Social Psychology, 45, 731-739.


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