Abstract realism in video games and aggression (Barlett & Rodeheffer, 2009)

I’ve read a post the other day about some observations on the importance of realism between North American and Japanese video game market. (I got another one) It’s one of the different priorities video game developers have. I could guess this could extend to the old media industry as well, of the few foreign television shows I watched (mainly my parents’ shows), the production quality is different and from a first impression, it seemed “inferior”. It’s something that I would investigate in the future.

Christophers Barlett and Rodeheffer have published a paper in Aggressive Behavior on the role of realism on video game violence effects on aggression (cognitive, affect and arousal). Mind you, the concept of realism is complex and multifaceted.


Previous research has shown that playing violent video game exposure can increase aggressive thoughts, aggressive feelings, and physiological arousal. This study compared the effects that playing a realistic violent, unrealistic violent, or nonviolent video game for 45min has on such variables. For the purpose of this study, realism was defined as the probability of seeing an event in real life. Participants (N574; 39 male, 35 female) played either a realistic violent, unrealistic violent, or nonviolent video game for 45min. Aggressive thoughts and aggressive feelings were measured four times (every 15min), whereas arousal was measured continuously. The results showed that, though playing any violent game stimulated aggressive thoughts, playing a more realistic violent game stimulated significantly more aggressive feelings and arousal over the course of play.

I just realized that for several months I haven’t watched television for an uninterrupted half-hour. I’m mostly on the internet keeping up with the news.

Whenever realism is mentioned in media or elsewhere (quite often by those lacking in gaming experience), it is often inferred to either its sensory meaning (i.e. graphical and auditory fidelity) or abstract meaning (i.e. likelihood of happening in real life). However, they neglected to consider both aspects. Their concern is genuine, people’s media experiences are influenced by degrees of realism and they can be immersed in these experiences as if they felt were real. Because of this immersive experience, it affects our behaviours, thoughts, feelings, and arousal. So the impetus of this study is to investigate abstract realism’s effect on aggression. Of the video games with high sensory realism, Half-Life 2 or the Crytek games come in to mind. As for abstract realism, any video games based on current events, for example Far Cry 2 where it draws real life elements like blood diamonds (it’s one of the few I know since I don’t often play FPS).  If you would please go to chapter 19 of this book for a review on video game realism, I would be very appreciative.


Participants: 74 university students somewhere in the Mid-West. Near equal gender ratio, average is 21.51 (SD =3.59), 72% are Caucasians 62% are not first-year students. Participants are given a chance to win a reward of a gift certificate to a local video game store and extra class credit.


Aggressive thoughts: the word completion task. Fill in incomplete words. It was administered four times, so the measure was split into four. So participants fill 24 different word fragments.

Aggressive feelings: the state hostility scale. It’s a 35-item, 5-point scale.

Trait Aggression: Buss-Perry Aggression Questionnaire. It’s a 29-item, 5 point-scale.

Physiological arousal: seven electrodes were attached onto the participant’s body. Physiological data was collected continuously, although data was analyzed on certain time points because any movement would create confounds in the data and that can be frustrating for the poor researcher doing the analysis.

Demographics: age, gender, ethnicity, etc.

Suspiciousness: because video game research is highly publicized, the researchers hide the true purpose from the participants until they are debriefed. 14 participants were found to be suspicious at debriefing. Their analysis has found that the suspicious participants’ data were not significantly different from the rest, so their data is included in the analysis. Perhaps, it’s time to dispense the need for such secrecy…

Video games used: they conducted a pilot study to determine which game to use in the primary study. So they recruited volunteers to find a game using a questionnaire to rate the games that is abstractly realistic vs. unrealistic, violent vs. non-violent, has third-person perspective gameplay and uses the same console.

Three video games were selected: Conflict Desert Storm as the realistic violent game, Star Wars Battlefront 2 as the unrealistic violent game, and Hard Hitter Tennis as the control nonviolent game.

There’s a flaw in their selection scheme in that pilot participants rated the games by watching a video of someone playing one of the video games, they did not play the video games. Why they decided on this is beyond me. Additionally, Star Wars Battlefront 2 didn’t have any depiction of blood (presumably fighting against the droids) and was noted, fortunately the pilot participants rated that the two violent games were not significantly different in terms of violence. One funny observation is that the researchers could not tell if their non-violent game is realistic because some of the questions seem inappropriate to ask (e.g. how realistic are the tennis rackets?), oh come on can’t you ask anyway or ask how realistic the game physics were?


Participants are randomly assigned to one of the video games. The realistic violent condition has 26 participants, unrealistic violent condition has 25 and the control condition has 23. Participants had electrodes to their bodies, they fill in a set of questionnaires, measuring for aggressive feelings, thoughts, and trait aggression. After a brief tutorial, they play their game for three 15 minutes play periods. In between periods, they fill the same set of questionnaires, measuring for aggressive feelings and thoughts for about 3 minutes. After the third play period, they had their electrodes detached, filled in the demographics, debriefed and were shown the way out.

Now I’m a bit worried about the procedure, since it’s a repeated design where, carryover or practice effects might be a factor in some of the measurements. The authors don’t believe to be a concern since participants weren’t given much time to remember the questionnaires’ content and that the word completion task was split into four unique pieces. An earlier study (Tamborini et al., 2004) did a similar procedure and I’m pretty sure that Barlett and Rodeheffer improved on that. Given the breaks between periods was short as a bathroom break and presumably the game is paused, it may be acceptable. Still, any breaks, even for a brief moment, to measure something would definitely have an effect. Perhaps, in future studies, we can use measures for aggression that doesn’t require breaks.


I decided to illustrate the results. They’re for illustrative purposes and they do not accurately depict the actual numbers. See the article itself. The points in each time period that have the same colour means they’re not significantly different. Those with a different colour means they are significantly different.

Figure 1: trait aggression, gender and aggressive feelings at time 1 were treated as covariates. Barlett and Rodeheffer argued after an initial increase in aggressive feelings, depending on the realism of a video game, it could either increase further and then stabilize or decrease slightly. However, given the results they have, this is more speculation that requires more work. The result support that realism affects us through a specific pathway, the emotional pathway (IMO), if we combine the results with physiological arousal, because highly immersive experiences lead to a high degree of involvement. So, in a sense, our emotional side of the mind is lead into believing a realistic artificial environment and reacts accordingly.

Figure 2: trait aggression, gender and aggressive thoughts at time 1 were treated as covariates. The reason that time 3 seem different from the rest of the data is due to the measure given was difficult for the participants to think for aggressive words. This is interesting, no differences in aggressive thoughts between the realistic and unrealistic violent video game. This is in line with previous research that cognition is primarily concerned with priming content irrespective of realism, so you can say the cognitive side of the mind can’t see the difference what’s real or not. Or simply put, violence is violence.

Figure 3: trait aggression, gender and heart rate at time 1 were treated as covariates. There seems to be some mismatch between the figure and the text, especially at time 3. So I’m not sure what to make of it. The authors found realism to have an important role in increasing arousal for the first 15 minutes of play, but it diminishes as it goes on. This effect is possibly due to getting comfortable with the graphics or a desensitization effect. A limitation is that they didn’t ask the participants to rate the game, so there’s no way of knowing if the game is boring or exciting as time goes on.

The authors concluded with the following: “In other words, these results suggest that if one plays a realistic violent video game, there will be an initial increase in aggressive thoughts and arousal and in aggressive feelings after 15 minutes, and those increases will be maintained or increased over the course of playing a realistic violent game.” I guess that if you have stubborn kids who want to play violent video games, it’s better to play Halo 3 than Modern Warfare 2. Maybe using parental controls for realistic games, such as changing blood colour or changing the gameplay slightly, say energy-based weapons instead of real guns?

Barlett, C.P., & Rodeheffer, C. (2009). Effects of realism on extended violent and nonviolent video game play on aggressive thoughts, feelings, and physiological arousal. Aggressive Behavior, 35, 213-224.


7 thoughts on “Abstract realism in video games and aggression (Barlett & Rodeheffer, 2009)

  1. After a bus ride home, I realized that video game realism has third “dimension” to consider: hyperreality and this was implied throughout many gaming blog posts (that I will have to wade through my browsing history). It’s strange how hyperrreality wasn’t considered in the study or slipped through my mind.

    • what do you mean by hyperreality?

      it sounds interesting, but I have no idea how you would measure s.th. like that – and I’ve never heard about any study regarding the phenomenon “hyperreality”.

      perhaps you can send me some articles or links regarding this topic

      By the way: Thank you for you really good summaries of these interesting studies concerning video games – well done.

  2. It would have been interesting if the researchers had compared groups between a violent game with “blood on” vs a group playing the same game with “blood off.” Assassin’s Creed 2 comes to mind, as a recent example. I suspect similar levels of “hostility” would have been measured, but it would still make for an interesting study …

  3. A fascinating discussion is worth comment. I think that you
    ought to publish more on this topic, it may not be a taboo matter
    but usually people do not speak about such topics.

    To the next! Cheers!!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com 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