To play or vent the anger: how goals determine the influence of violent videogames on aggression (Denzler et al., 2012)

“Shoot, shoot, dodge, shoot weakspot, boss is killed, objective complete… go to bathroom, pull pants, sit on toilet and poop.”

During my stay at Ohio State, I was intrigued by Roy Baumeister and company’s Strength Model of Self-Control, their work became one of the theoretical bases for my master’s thesis. I skimmed a recent publication of Masicampo and Baumeister’ s who examined how plan making can attenuate the nagging mental strain of unfulfilled goals and fulfilling whatever goals in the experiment seem to liberate your mind for whatever new tasks is given. If those goals went unfulfilled, it would’ve been littering in your mind or in my case, gnawing my attention away.

Markus Denzler (University of Amsterdam) and colleagues examined whatever fulfilling a goal of venting anger through watching, playing a videogame can inhibit aggressive thoughts. I was hoping they examined how objective completion (say kill 50 enemy soldiers or aim for the high score in 15 minutes) might have attenuate aggression, something that would explain Autumn quarter’s results.


Generally, the accessibility of goal-related constructs is inhibited upon goal fulfillment. In line with this notion, the current studies explored whether violent computer games may reduce relative accessibility of aggression if the game involves the fulfillment of an aggressive goal. Specifically, in Study 1, participants who watched a trailer for a violent computer game that fulfilled the goal of venting anger showed less relative accessibility of aggression compared to participants who watched the trailer without goal fulfillment. In Study 2, actually playing a violent computer game to vent anger also decreased the relative accessibility of aggression compared to a control condition in which the game was played without such a goal. Lastly, in Study 3, the relative accessibility of aggression was reduced after playing a violent computer game for participants who reported a high general tendency to vent their anger.

The problem of being a graduate student is that there are so many goals that you lose track of them, but only for them to resurface at the worse possible moment.

The current dominant theoretical model of how violent media increases aggressive and violent thoughts is the priming principle where one exposure to one concept activates other concepts in a sort of cognitive map or schema. For example, when I say the Eiffel Tower, you think of the French, champagne, berets and all of those French sorts of things. The same applies to violent videogames: kills, frags, shotguns, blitzkrieg, etc.  The authors argued for a dynamic model to this priming principle and that aggressive thoughts might be moderated by goals or motives of playing violent videogames.

The authors related this motivational principle of goal fulfillment and the Zeigarnik effect. A classical example of the Zeigarnik effect is how waiters seem to remember your order and the seating arrangement during your entire meal as it is their objective to serve you in good service, but as soon the bills are paid when he or she is done with you, they don’t recall very well afterwards. The authors argued that playing violent videogames for the sake of playing should increase aggressive thoughts and these thoughts are brought to the foreground of consciousness much louder before you accomplish your goals. However, the authors asked an interesting question. Much like the waiter who is finished with a table, what if a player has accomplished a goal, say venting anger? Their aggressive thoughts should recede from their consciousness if mission accomplished, but they will stay when these goals are not accomplished yet or unattainable. I must agree with this line of thought, although not with aggression, but through my OSU! experiences whenever I broke a combo. I try to go back to grading papers, but the lingering thoughts and strategizing brought me back to playing OSU! once more just to try it out or perhaps thinking the next time I will get the perfect combo.

Quite interestingly, Bushman & Gibson (2010) conducted a related study examining rumination. Participants who were told to think about their gameplay of a violent videogame overnight were more aggressive behaviourally than those who were not told to do so. The act of ruminating actions involving violence distracts the mind and our behaviours towards whatever is in mind.

This is quite an important methodological consideration as I read this article, it makes me wonder how participants in past studies were treated when playing a videogame: were they told they would play for a certain amount of time? How did the researchers stop game play? Did they interrupted the participants in the middle of their game or do participants were given advance warning or saw the clock ticking to the end? What were they told to do? Just play the game or play the game for the high score?  Not many researchers reported this kind of behaviour as it is not something that could be important to think about.

To demonstrate their line of thought in one of Denzler’s studies, participants were asked to imagine taking revenge against a person who provoked. What they found is that aggressive thoughts are decreased only if they got their revenge to the specific person, but not if they vented their anger to a punching bag, an unrelated object or “person”. That sounds logical, but then the problem remains is that the act of aggression was already completed. Furthermore, the means for goal fulfillment is another important component. They argued that anger in general comes from life frustrations (e.g. work, relationship problems, reality knocking, etc.) and there are different means that helps achieve whatever goals one has set. Videogames, they argued, can be a means for a variety of goals from relaxation, achieving something palatable to venting anger. If one wants to hurt someone, a violent videogame can supply a means to hurt someone. If someone wants to hurl a pie to a politician, I think an online search for such games would be appropriate.

Study 1


Participants: 47 participants, 24 females and 23 males. No other demographical information aside from the clue that they are Germans.


Mood: one-item on a 7-point scale: “How do you feel?” from very bad (1) to very good (7). Whatever happened to using the PANAS?

Aggressive cognition: The Lexical Decision Task (LDT) where participants are tasked to determine whether a string of letters is a word or a nonword as quickly as possible. The participants were randomly present to 7 aggressive words (e.g. Strafen “to punish”, angriff “attack”),  7 neutral words (e.g. Schlafer “to sleep”) and 14 nonwords (Grumpf). Each word appear after the participant’s response and they stay until the participant responded.


Participants were tested individually and were lead to believe that they are participating in three unrelated studies. First, the complete the pre-test mood question. Then they were told to imagine a person, the same sex as the participant, who caught his or her partner cheating with his or her best friend. They were told to experience the thoughts and feelings of the person as the story unfolded. The story differed by gender, such as the female version where the person runs away without them noticing and is thinking her sharing intimate information with her best friend and such.  Participants wrote down what the person is thinking, feeling and behaving in the near future.

Afterwards, they went on to complete the lexical decision task under the guise of a reaction time task. After that, they answered a second mood question. Participants are then led back to hypothetical story, where the person goes home and play a violent videogame (i.e. Battlefield 1942). The participants are randomly given one of two story versions: In the goal fulfillment condition, the person is playing to vent anger and that the person felt better. In the no-goal fulfillment condition, the person is playing to distract themselves and didn’t feel any better. To facilitate the story, the participants watch an 80sec trailer of the videogame. After that, they completed another lexical decision task and then a post-test mood question.


They analyzed the data using a mixed model ANOVA examining the two LDT and goal fulfillment conditions with reaction time to the words as their dependent variable for aggressive cognitions. They found that, through pairwise comparisons, there are no differences between goal fulfillment conditions in the first LDT, since this was given before the goal fulfillment condition. They did found differences in the second LDT, which was after the goal fulfillment, where does in the goal fulfillment condition had lower accessibility to aggressive cognitions than those without the goal fulfillment.

Their analyses on participants’ mood at pre-test, after the scenario/before the goal fulfillment (2nd take) and after watching the trailer revealed that participants felt worse in the 2nd take than the pre-test and their mood after watching the trailer did not feel any better or worse at all, not even group differences.

Study 2

Study 2 is an extension of study 1 with some adjustments.

Participants: 28 participants, 14 females and 14 males.


Participants were told to remember a situation in their own life where they were very angry at someone, which brings up some angry feelings and thoughts. They wrote down these feelings and thoughts and answered the mood question which is followed by an LDT with different words from the first study. Participants get to play a shooting game (they called it an FPS, but I disagree) which was taken from another study. Play time is 4 minutes, which is rather awkwardly short. Participants were randomly told to play the game either to vent their anger or to play the game. After that, a post-test LDT and mood question.


They analyzed the data using the same statistics as in the first study. They found that participants in the goal fulfillment condition displayed lower aggressive cognition accessibility than those in the no-goal condition. As for mood, they found no differences between conditions and that their mood worsened over the course of the experiment, so the authors wrote in a way that the participants felt better after remembering an angry event than after playing videogame.

At this point, I felt that these experiments were just being too obvious to the participants and they were responding to the experimenter’s expectations. Furthermore, I am bothered by the small sample size which makes nervous. Fortunately, the authors had the foresight to examine this relationship in a more naturalistic way.

Study 3

Study 3 is a replication of study 2 with some modifications.

Participants: 46 participants, 25 females and 21 males.


Anger: they used the Anger-out subscale from the Anger Expression Inventory. The subscale is an 8-item 4-point frequency Likert-scale.


Before the start of the experiment, participants filled out the Anger Expression Inventory. They completed an LDT which is used as baseline. Then, participants were told to remember a situation in their own life where they were very angry at someone, which brings up some angry feelings and thoughts. They wrote down these feelings and thoughts and answered the mood question which is followed by an LDT with different words from the first study. Participants get to play that same shooting game . Play time is 4 minutes, which is rather awkwardly short. This time, participants were told to play the game. After that, a post-test LDT and mood question.


To test the anger-out scores, they analysed the data through multilevel data analysis. Their results indicated that higher Anger-out scores is associated to higher aggressive cognition accessibility.

In terms of the three blocks, from baseline, after writing the angry story and playing a violent shooter videogame, they found that among those with low anger-out scores, their scores increased after the angry story and stayed relatively the same after playing the violent videogame, however they’ve noted the mean score were lower, but it was non-significant or rather a statistical trend. Among participants with high anger-out scores, their trajectories are the same with their low anger-out counterpart with the exception that the after-videogame play aggressive cognition accessibility scores were significantly lower than the anger story, but not baseline or rather in a sense, their scores returned to baseline levels.

As for mood, ANOVA results with anger-out as a covariate revealed only main effects for the block conditions. Participants felt worse after the angry story, but their mood became better after the videogame play. The authors noted that they are reluctant to explain why this occurred, they noted that participants were feeling cheerier as they approached the end of the experiment. I don’t know about their participants knowing the course of the experiment, but I have some of my participants not knowing what they were supposed to do, even though they read (or should I say skimmed) the consent form.

The authors conducted a mini-meta-analysis of their three studies because not all of their studies yield similar results. Skipping to the last lines, the meta-analysis showed that aggressive cognition accessibility is reduced following violent videogame play when it serves as a cathartic goal. Furthermore, they added that the effect size is similar to prior reports for increases in aggressive cognitions.


An interesting thought I had is the connection between this study and self-determination theory. Those with intrinsic motivation tend to create and engage in their goals by their free will. On the flipside, those with extrinsic were pressured by external forces and the goals they have were not of their own (which from this theory explains their aggression). I have not a clue for an example on how these two theories would weave out.

The authors noted about other goals, such as relaxation or entertainment, and whether these semantically different goals would have the same effect as the cathartic goal. Well, I think I may have stumbled upon this kind of data, except that I told all of my participants to kill or freeze as many opponents as possible in a time limit handled by the videogame’s timer. Afterwards, my RAs and I noted their scores, I am not sure if my participants would though in succeeding their goal, but at least it is fulfilled. But the goal factor might have explained the non-significant results.

Do note that whilst aggressive cognitions are quelled away, the bad mood still lingers, even after playing a violent shooter videogame. The authors argued for catharsis conceptual models to distinguish the emotional and cognitive pathways and how they affect behaviours, especially aggressive behaviours.  The non-accomplishment of such goals could lead to rumination and this could lead to increases of aggression, in the particular case of violent videogames.

Another concern I have is to investigate further is to identify problematic goals. Whilst goals stated in self-report indicate entertainment, relaxation or competitive fun, I am concerned about the in-situ situational goals that might affect which is in contrast to goals reported after a temporal distance away from gaming time. It is possible, given my experience, that some might have goals that cannot be completed or that decreases one’s self-efficacy. One prime example is getting hard-to-get achievements/trophies or being obsessively and stubbornly passionate. In my case, getting through a beatmap without any mistakes, on hard or insane.

The authors noted the target of such aggression might be a concern as well. If a vexed player can vent their anger on a specific target, such as participants who did not show up in my study, and symbolically harm them in a videogame, it would be interesting to see if it does reduce aggressive cognition. On the other hand, they noted that a vague target, such as an institution, is less likely to be fulfilled and thus less likely to quell hostile thoughts.

One external validity issue is whether people who play videogames do so with a goal in mind, or how often do gamers start playing with goals in mind. Perhaps they see a gamepad and started playing, like some sort of delicious distracting snack. This could pose some problems and given from the study’s implications, increases in aggression might result. Furthermore, telling participants to “play the game” is not something that could convincingly be construed as a goal itself, participants might anchor towards another goal of some sort: achieve high score, play the game thoughtlessly for the project, etc.

The take home message is that having goals, especially of an aggressive nature, while playing a violent videogame and achieving them in good order would serve well as a cognitive catharsis, but we still have the emotional issue to deal with.  Although, I don’t think anyone can run their behaviours on emotions alone.

Denzler, M., Hafner, M., & Forster, J. (2012). He just wants to play: How goals determine the influence of violent computer games on aggression. Personality and Social Psychology Bulletin , 37 (12), 1644-1654. URL DOI: 10.1177/0146167211421176



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