Good Ash vs. Bad Ash: How contexts affect aggressive behaviour and cognition (Gitter et al., 2013)

When I saw that article popped up back in May 2013, I had the feeling readers were anticipating my review of this study. Was I right?

Seth Gitter (Auburn University), Patrick Ewell (University of Alabama), Rosanna Guadagno (NSF), Tyler Stillman (Southern Utah University) and Roy F. Baumeister (Florida State University) have published an interesting study in Aggressive Behavior, examining context of violent behaviours and its effects on aggressive behaviours and cognition.


Previous work has shown that playing violent video games can stimulate aggression toward others. The current research has identified a potential exception. Participants who played a violent game in which the violence had an explicitly prosocial motive (i.e., protecting a friend and furthering his nonviolent goals) were found to show lower short-term aggression (Study 1) and show higher levels of prosocial cognition (Study 2) than individuals who played a violent game in which the violence was motivated by more morally ambiguous motives. Thus, violent video games that are framed in an explicitly prosocial context may evoke more prosocial sentiments and thereby mitigate some of the short-term effects on aggression observed in previous research. While these findings are promising regarding the potential aggression-reducing effects of prosocial context, caution is still warranted as a small effect size difference (d = .2–.3), although nonsignificant, was still observed between those who played the explicitly prosocial violent game and those who played a nonviolent game; indicating that aggressive behavior was not completely eliminated by the inclusion of a prosocial context for the violence.

I’m heading out for NCA at Washington D.C., I will appear when I am supposed to.

A frequent criticism of violent videogame research I hear is that experimental results were conducted without context, participants were just thrown into the game and told to shoot up some people. It does not take the narrative into account that the player-character is the protagonists and that their acts are morally justified, hence ecological or external validity is low.

The authors argued that eliciting prosocial thoughts along with the violence in videogames would mitigate its effects. A storyline as a means to provide a prosocial mindset would lessen aggressive behaviours. Their theoretical perspective is from my advisor’s General Aggression Model where in short violent stimuli prime people towards aggressive thoughts, behaviours and feelings. The converse is true where prosocial stimuli can prime prosocial thoughts, behaviours and feelings. In videogames, violence is often justified because they are the hero, they have the moral/prosocial license to kill. Although a meta-analysis (Anderson et al., 2010) found no differences in aggression whether participants played as a hero vs. a criminal. The authors argued that the meta-analysis did not account for the complicated motives of player-characters, say a criminal robbing for the poor. Hence, the authors argued that leading participants to interpret the aggressive acts of a character in a prosocial manner would lessen priming of aggression. Second, priming of aggression and prosocial are usually not independent of each other and can be activated at the same time, so the authors argued that because more cognitive concepts are activated, it is not just the aggression is on our mind, but also prosocial ones which would decreases the likelihood of aggression.

All is well said for psychologists, but I am left unsatisfied with their literature review because of my habit of collecting articles and my experiences in communication science. So here are a few things they miss. First, I am pleased that they mentioned Leonard Berkowitz’s research, I wish they would look at what communication scholars ran with his research. W. James Potter (UC Santa Barbara) conducted a lot of research on context, justification, viewer interpretation among other things in media (mostly television). Justification of violence is certainly known among media effects scholars and unfortunately not much research has arisen since… a good while, but Stacy Smith’s (USC Annenberg) work would be a good start for our authors on further theorizing their research question. Second, a counterpoint to this study is moral disengagement in videogames as exemplified by Hartmann & Vorderer (2010). What are the effects of being a given a just and moral cause to harm other humans, unlike the zombie game in the current study? Third, the cognitive comprehension of videogame narrative can be a good avenue to explain how gamers interpret their play experiences of which influence their aggressive or prosocial behaviours towards others. Potter & Tomasello (2003) is probably a good start. To my knowledge, however, there is no one who studies comprehension in videogames. For anyone with a strong background in cognitive science and has an interest in narrative should contact David Ewoldsen (Ohio State University), get started with the Landscape Model by Paul W. van den Broek or the situation models by Gabriel Radvansky.

Study 1

The first study examined the effects of prosocial vs. morally ambiguous contexts on aggressive behaviours.

Participants: 100 male undergraduate students from some introduction to psychology somewhere in the warm parts of the United States, I think. 19 students were taken out of the analysis because they were suspicious of the experiments. Why only male participants because they are familiar with videogames, that is all and I heartily agree from a methodological standpoint.


Trait aggression: Buss & Perry Aggression Questionnaire which assess for individuals’ personality trait of aggression. It is a 29-item scale on a 5-point characteristic scale.

Videogame experience: participants reported the number of hours of videogame play per week and listed their three top favourite videogames.

Aggressive behaviour: the Competitive Reaction time task. Participants were told that they would be competing with another student on a reaction time task. The participants can set a noise blast ranging from 0db to 105 db and the duration from 0 sec to 5 sec (as I vaguely recall) for their partner. Whoever loses a trial is blasted by the noise set by the winner. The number of trials is 25 and they were fixed in that the participants wins 13 trials and loses 12 trials.

Videogame used: The Evil Dead series. Evil Dead: Regeneration is used as the prosocial game as the player-character is tasked to protect others from zombies and helping other characters. Evil Dead: Fistful of Boomstick is used as the morally ambiguous game as the player-character is tasked to kill as many zombies as possible. The authors termed Boomstick as morally ambiguous because of the possibility that players would interpret it as the player-character having prosocial motives. A third control game is Tetris Worlds because the Evil Dead games are both violent.


Participants were given a tutorial of the competitive reaction time task, in an effort to reduce the time delay between videogame play and assessing aggressive behaviours. After the tutorial, they were given a tutorial on how to play one of the videogames and played for 15 minutes. After that, they competed on the noise task and then completed the questionnaire.


The authors checked whether trait aggression and videogame experience might have a role in the effects of violent videogame and aggression and found none.

The authors analyzed aggressive behaviour based on the first trial of the competitive reaction time because it is a pure measure of aggression as subsequent trials is more on reciprocating behaviours. Their one-way ANOVA revealed that those in the morally ambiguous game were more aggressive (M=6.69, SD=2.11) than those in the prosocial (M=5.48, SD=2.15) and control (M= 5.07, SD=1.54) game. The prosocial and control were not significantly different from each other. The authors did examined the data in different ways, such as aggression across all 25 trials, after win trials or loss trials. They found that those in the morally ambiguous game were more aggressive after a loss trial. Across all 25 trials, those in the morally ambiguous game were more aggressive than those from the control game, but not from the prosocial game. The prosocial game was not significantly different from the control game.

Study 2

The Evil Dead games were equally violent with the exception of the player-character’s motives. The authors examined whether the decrease in aggression is because there were more prosocial thoughts elicited in players’ thoughts.

Participants: 132 undergraduate students , one participant was removed from the analysis due to suspiciousness. 53% of participants are male.


Word fragment task: Participants were to fill a list of 98 incomplete words (e.g., E N G _ A E) in a 5 minute time limit. 30 words could be completed either in hostile or neutral wording and another 30 could be completed in either prosocial or neutral wording. As the task assessed for the salience of both aggressive and prosocial thoughts at the same time, the authors opted to analyze the data in terms of the proportion of completed prosocial or hostile words to the total number of words completed.

Trait aggression: Buss & Perry Aggression Questionnaire which assess for individuals’ personality trait of aggression. It is a 29-item scale on a 5-point characteristic scale.

Videogame experience: participants reported the number of hours of videogame play per week and listed their three top favourite videogames.

Videogames used: The same Evil Dead games, and Snowboard Supercorss (SSX).

Procedure: Participants play one of the videogames for 15 minutes. They were also given a tutorial. Then they were given 5 minutes to complete the word fragment task, then completed the questionnaires.


The authors found that the videogames were not equal on several characteristics, namely difficulty, enjoyability, excitement, and pacing as covariates in their analyses. Frustration was also added as covariate when analysing the prosocial results.

Their ANCOVA analysis for prosocial thoughts revealed that participants across all games differ from each other, in the prosocial game (M=0.166, SD=0.06), in the morally ambiguous game (M=0.138, SD=0.055), and in the control game (M=0.13, SD=0.06). Their ANCOVA for hostile thoughts revealed non-significant differences, although the means were what we would expect.


The take home message is that a prosocial context in a violent videogame can decrease aggressive behaviours and make more prosocial thoughts more salient and accessible in players’ mind.

The authors explained why those in the morally ambiguous game completed a higher proportion of prosocial words than those in the control game may be due to what happens in the Fistful of Boomstick as participants may have interacted with non-player characters in the game, even though the game did not explicit show any prosocial motives, the participants might have inferred them.

The authors listed some limitations and interpretive issue. First, the prosocial context of the violent videogame did not eliminate the media violence effects, there is reduction nevertheless. Second, they pondered whether previous videogame studies would have reduced effect size because of the morally ambiguous context from the videogames they used. Perhaps, the effect sizes would be larger by eliminating any prosocial contexts. The authors noted that they not examined the mediation of the relationship between videogame play context and aggressive behaviour through prosocial/hostile cognition because the direct effects of the independent variable on the dependent variable could be reduced. Third, the long terms should be examined in future studies.

I have some considerations stemming from this study. Let us consider the prevalent narratives of military-themed videogames, although I do not know any content analyses on videogame narratives, but the anecdotal common narrative is centered around protecting one’s ingroup be it a country or company, it is always defending, seeking retribution or framing their violent actions as just and good. Is there any popular videogame narratives where the player-character is truly cast as immoral and violent? First, how are the outgroup, the enemy, are portrayed and how would players behave towards the enemy? Then we should be examining the issue of dehumanization and behavioural outcomes between ingroup members and outgroup members. Fortunately, John Velez and colleagues (2012) showed that cooperating with an outgroup member can reduce aggression. Second, priming a prosocial context in violent videogames in particular for a realistic and military videogame might elicit some positive evaluations towards the military, Ruth Festl and colleagues (2013) found no such relations of positive towards the military and videogame experience among German players, but who knows with the Americans. It would be interesting to see how priming honour among participants would affect their in-game behaviours and in real life as well. Third, I’d like this theoretical conflict (if it is) between this study’s findings and the issues of moral disengagement found in by Hartmann and Vorderer (2010) resolved.

Gitter, S. A., Ewell, P. J., Guadagno, R. E., Stillman, T. F., & Baumeister, R. F. (2013). Virtually justifiable homicide: The effects of prosocial contexts on the link between violent video games, aggression, and prosocial and hostile cognition. Aggressive Behavior, 39 (5), 346-354. DOI:10.1002/ab.21487


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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s