Overview   •   •   •   Diploma thesis   •   PhD thesis   •   Cortex   •   Press

The Effects of Displayed Violence and Game Speed in First-Person Shooters on Physiological Arousal and Aggressive Behavior.

+ awarded the 2012 Paul F. Lazarsfeld-stipend of the German Communication Associaton (DGPuK)

Below, you find a short abstract of my diploma thesis, the full thesis pdf file, and the violent-free toy gun mod for Unreal Tournament 3 that I created for it.


Many studies have been conducted to examine the effects of displayed violence in digital games on outcomes like aggressive behavior and physiological arousal. However, they often lack a proper manipulation of the relevant factors and control of confounding variables.
In this study, the displayed violence and game speed of a recent first-person shooter game were varied systematically using the technique of modding (see figures 1 and 2), so that effects could be explained properly by the respective manipulations. Aggressive behavior was measured with the standardized version of the Competitive Reaction Time Task or CRTT (Ferguson et al., 2008): The participants played 25 trials of a reaction time game against a fictional opponent (actually a computer), in which they had to press the space bar as fast as possible after hearing a sound signal. The loser of a trial would hear an obnoxious noise blast whose volume and duration was set by the winner. The volume and duration settings of the participants are considered the measures for aggressive behavior.
Physiological arousal was operationalized with four measurements: finger clips for galvanic skin response (GSR) and heart rate (HR), a Wii balance board the participants sat on for body movement, and sensors on mouse and keyboards to measure applied force. A total of N = 87 participants played in one of four game conditions (low- vs. high-violence, normal- vs. high speed) while physiological measurements were taken. After play, their aggressive behavior was measured with the CRTT.

The results of the study do not support the hypothesis that playing violent digital games increases aggressive behavior, neither for the CRTT's average volume nor for duration settings. There were no significant effects of displayed violence and game speed on GSR and HR, but with a higher game speed, participants showed less overall body movement, F (1, 80) = 10.47, p < .05, ω =.31, most likely to meet the game's higher demands on cognitive and motor capacities. Also, higher game speed and displayed violence caused an increase in applied force on mouse and keyboard, F (1, 80) = 14.24, p < .05, ω = .35. Previous experience with digital games did not moderate any of these findings.
Moreover, it provides further evidence that the CRTT should only be used in a standardized way as a measurement for aggression, if at all. Using all 7 different published (though not validated) ways to calculate levels of aggression from the raw data, "evidence" was found that playing a violent digital game increases, decreases, or does not change aggression at all.

Thus, the present study does extend previous research. Firstly, it shows the methodological advantages of modding in digital game research to accomplish the principles of psychological (laboratory) experiments by manipulating relevant variables and controlling all others. It also demonstrates the test-theoretical problems of the highly diverse use of the CRTT. It provides evidence that for a meaningful interpretation of effects of displayed violence in digital games, there are other game characteristics that should be controlled for since they might have an effect on relevant outcome variables.
Further research needs to identify more of those game features, and it should also improve the understanding of the different measures for physiological arousal and their interrelatedness.

  Download Full Thesis (5 MB)


This package includes several mods for Unreal Tournament 3 that I used in my diploma thesis, of which the first 3 I created myself. You will find instructions on how to install it in the *.rar archive.
ToyGun - A nerf gun that shoots tennis balls (see image).
FlakGun - Modification of the Flak Cannon, so it would have the same functions as the ToyGun (but still look like the Flak Cannon).
Death Fading - Bodies no longer explode, but become spectrally transparent once they "die" (see image).
UT3 Stats Logging - Writes an exhaustive logfile for every match.
UT SpeedMod - Allows you to set the game speed from 50% to 400%.
Language - German language settings that replace violent words with competetive ones.

  Download Mods (11 MB)

cite as

Elson, M. (2011). The Effects of Displayed Violence and Game Speed in First-Person Shooters on Physiological Arousal and Aggressive Behavior. Diploma thesis. University of Cologne.