Devilly, G. J., Unsworth, G. , Ward, T. (in press) The effect of playing violent videogames on adolescents: Should parents be quaking in their boots? Psychology, Crime & Law.
This article came when gamepolitics.com found it in an australian news website. So I went and asked for a copy of the article from one of the authors. Unfortunately, I cannot distribute the copy freely without his permission or the journal’s. A news site gives a summary of the study.
This is my first blog and therefore my baptism of blogging which would be messy. So what I would be writing will be just bad and hopefully improve over time.
The authors argued that the instrument used in previous studies are questionably reliable and argued the use of “reliable change index” which I don’t get and will need some further research. But it is argued that it is psychometrically reliable based on previous research and it will be the first time in using the RCI for this kind of study.
They also discussed the General Aggression Model (GAM) and pointed out two problems to the GAM: (1) personality, it is argued that personality is relatively stable and according to my textbook, Anderson et al. might support the learning or conditioning perspective of personality. it would be interesting to see whether trait perspective like the five-factor model of personality have a say in VG research. (2) “…the model does not consider other learning experiences which may be equally powerful and yet provide a less aggressive alternative to conflict resolution.” I might add whether players learn and play in terms on the rules and laws of the game and apply these analytical skills to other situations. For example, “Doom” does not have jump, the ability to look up or down, or intelligent A.I. Although, VG are becoming more realistic they are still bound by rules that players would eventually learn and exploit.
Their hypothesis is interesting in that individuals self-regulate their emotions and arousal level, much like self-control or a having an amount of energy for self-control based on two temperaments, labile and stable. I have never thought of that and i should read the GAM, if it mentions self-regulation. Implications are important because impulsive individuals may run into a lot of trouble and the extreme form of it is antisocial personality disorder and other personality disorders.
The sampling of participants is peculiar, restricting participants to those who played Quake 2. Ludologically(? must ask Jesper Juul) speaking, playing Quake 2 and an equivalent FPS shouldn’t pose some problems, except differences with the gameplay and the setting of the game, although it is unknown whether they may affect differences in results. This would be an interesting research question whether the structure of a game is the same but the situation is different, i.e. pre-exposure of a one gaming situation then testing to a different game. So they learn how to operate fundamentals, but they would run different mazes. One might ask whether developing a proficiency test for a gaming genre may be useful for research.
Continuing on about the participants, what about the gaming-learning context: playing in single-player mode is different from mutiplayer, reactions and such differ in these two modes. Players had played the game before so from the start of the single-player campaign may be a bit boring for some since they know what will happen in the level. 20 minutes of gameplay is somewhat short, but it is logical in order to keep time factor controlled. I’d prefer a free-run experiment to control other individual differences like arousal or boredom.
What’s interesting is the attempt to assess participants’ state during play using “Articulated Thoughts in Simulated Situations (ATSS) Paradigm”, basically players say their thoughts out loud druing play. Other studies have done unobstrusive observations before (see “handbook of video game studies”), they found no aggressive behaviours manifested during play. Another interesting thing is that they are measuring ‘anger’ and they said that is not a measure of aggressive behaviour, but they did argue that anger is relevant to aggression. Sounds counterintertuitive, but I can’t dismiss it…
I’ll skip the stats, it’s beyond my knowledge.
The authors acknowledge that state affect or emotion may not be good to measure aggression and proposed a behavioural measure instead of angry affect. Nevertheless, I believe they should continue, what if we attempt to manipulate affect during play and a control group doing an equivalent task. Should the play group be more angry than the control group. Or do the violent game group more angry than the non-violent game group?
Well I guess that’s all I could think of, but I’ll add more as soon my brain process the info some more.
(to authors: i may have omitted some portions of your study, but I think I’m bordering on copyright violation and/or I have nothing to comment on it.)
Unsworth, G., Devilly, G. J., & Ward, T. (2007). The effect of playing violent video games on adolescents: Should parents be quaking in their boots? Psychology, Crime & Law, 13(4), 383.