Don’t give Kratos your videogame controller: Trait anger and aggressive behaviours (Engelhardt et al., 2011)

This article published in Aggressive Behavior caught my attention when I ask myself if they were re-inventing the wheel. After a brief literature review, their study was most similar to Giumetti & Markey (2007), although Christopher Engelhardt, his advisor Bruce Bartholow and Scott Saults (University of Missouri) used different videogames and aggression measures.


Although numerous experiments have shown that exposure to violent video games (VVG) causes increases in aggression, relatively few studies have investigated the extent to which this effect differs as a function of theoretically relevant individual difference factors. This study investigated whether video game content differentially influences aggression as a function of individual differences in trait anger. Participants were randomly assigned to play a violent or nonviolent video game before completing a task in which they could behave aggressively. Results showed that participants high in trait anger were the most aggressive, but only if they first played a VVG. This relationship held while statistically controlling for dimensions other than violent content on which game conditions differed (e.g. frustration, arousal). Implications of these findings for models explaining the effects of video games on behavior are discussed.

Well I am done with my Master’s Thesis and the resulting fallout is my aimless wandering of the halls looking for my determined self.

Trait anger is defined as an emotional predisposition where responses lean more towards aggression than other types of responses. Anger is posited to be the result of a general negative mood and thus leading towards aggressive tendencies. Under cognitive theories, bringing up angry emotions would bring up its closest conceptual relatives, aggressive thoughts and behaviours. It is not necessarily through anger that elicits aggressive behaviours, it could be any of one them that could prime the other two.

The authors proposed that angry individuals playing violent video games would be more aggressive. They reasoned that as violent videogames prime aggressive thoughts, emotionally angry individuals would most likely to act more aggressively because it is the easiest behaviours they can use or rather behaviours most salient since both anger and aggressive thoughts are turned on in their consciousness.


Participants: 83 undergraduate students, average age is 19 (SD = 1.0), 21 females and 62 males. 6 participants were dropped from the analysis because of suspiciousness or being outliers.


Trait Anger: the anger subscale from Buss and Perry’s Aggression Questionnaire. It consist of 7-items on a 7-point Likert agreement scale.

Game ratings: participants rated the videogames they played under a 7-point agreement scale. There are 6-items assessing for frustration, excitement, engagement, interesting, arousing and violence.

Aggressive behaviour: The competitive reaction time task. Participants were to compete with “another participant” on a reaction time task. This task consisted of a single trial where participants were told they were evaluated on how fast they can recognize an image, and how accurate they identified the colour of the image. And at the last minute, the participants were told that they would be evaluated on how fast they set their noise levels, from 0 to 10, the higher the number, the louder and this is operationalized as aggressive behaviour. That is quite an interesting twist on this often used aggressive behavioural measure. I suppose that telling the participant this last criterion would leave them little time to think and therefore acted on their predispositions.

Videogames used: The violent videogames used are Mortal Kombat vs. DC Universe, Resident Evil 5, Killzone 2, F.E.A.R. 2: Project Origin, and Call of Duty: Modern Warfare 2. The non-violent videogames used are MotorStorm, NCAA Basketball 2009, Sid Meier’s Civilization Revolution, Little Big Planet and Ferrari Challenge.

Procedure: Participants completed the trait anger questionnaire. Then, a whole cover show was made to make the participant believed that another student is also participating. tl; dr. Participant then played their randomly assigned videogame for 20 minutes. Then, they completed the competitive reaction time task. Then, they completed the game rating and then debriefed.


Participants found the violent videogames to be more frustrating (3.7 vs. 2.8), exciting (5.0 vs. 4.2), arousing (3.7 vs. 2.8) and violent (5.5 vs. 1.2) than the non-violent videogames. The authors took these differences as covariates in their analyses, but none were significant, only frustration was marginally so at p = .08 which also might have been due to chance.

The analysis was conducted through a hierarchical regression. The regression analysis revealed that gender is a significant factor: men (M = 5.3) are more aggressive than women (M = 3.8). Trait anger as an independent factor was not a significant factor, but it becomes significant as an interaction with videogame condition when entered into the analysis.

Looking at the direction of this interaction, the authors found that angry individuals who played a non-violent videogame seemed to give quieter(ish) noise than their less angry peers, but the significance value is p = .06, so this may have been due to chance (by convention). Angry individuals who played a violent videogame do give louder noise than their less angry peers. The authors contend that angry individuals are more susceptible to video game content and thus their emotions fluctuate much more than others. Trait arousability might explain the study’s results and the direction someone with high trait anger would behave to different situations. IMO, it is perhaps they are more reactive to more sensory input that calls for further investigation into what other dispositions that goes along with trait anger, perhaps neuroticism or psychoticism.

On the other hand, looking at the averages of this interaction, the differences are not significant where the angry participants who played a violent videogame gave louder noise (M = 5.7, SD = 2.4) than the less angry participants (M = 4.3, SD = 2.5), although it is marginally significant at p = .09.


Their results are full of marginal significances, but two were significant of which are the videogame and trait anger interaction effects and the simple slope analysis in the violent videogame condition as a function of trait anger. An analysis from 77 participants… did they ran out of time to test more participants? I am taking these results with two tablespoons of salt and melted butter on corn. Nevertheless, it is rather a nice replication of Giumetti & Markey (2007).

It’s okay… my results are full of marginal significances and one significant main effect.

Engelhardt, C. R., Bartholow, B. D., & Saults, J. S. (2011). Violent and nonviolent video games differentially affect physical aggression for individuals high vs. low in dispositional anger. Aggressive Behavior, 37 (6), 539-546. DOI:10.1002/ab.20411


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