Differences in video game interface, content and context in relation to presence and hostility (Eastin & Griffiths, 2006)

Whenever I read an article on my computer screen, it takes more effort and time than to read it on real paper. Also, more ideas and questions pop up when reading in a comfortable position than sitting in front of a PC. Well I guess the computer revolution doesn’t eliminate the need for paper. Oh, if anyone knows a lightweight gadget and where the screen isn’t glaring like a television screen. Please leave a comment.


Investigating male game players, this study explores how game interface (virtual reality [VR] and standard console), game content (fighting, shooting, and driving), and game context (human and computer competition) influence levels of presence and hostile expectation bias—the expectation others will think, feel, speak, and act aggressively during social conflict. In addition to game interface and game content influencing hostile expectations, significant interactions were detected for hostile expectations. Presence, although not as predicted, also significantly differed across game interface and game content. Through the development and testing of each gaming experience, this study demonstrates that simply testing violent and nonviolent game situations underestimates the complexity of contemporary video-game play.

“I’m playing first-person shooters, you don’t see me shooting people after I played.” That’s a common counter-argument towards anti-videogames advocates. Well, that’s good against a certain audience, but not for others (i.e. researchers) because that’s just using an extreme example rather than a benign one and violent video game opponents would still argue that it has an negative effect anyways. However, that statement is related to Eastin and Griffiths’ study and it answered whether first-person shooters “train” individuals to be more aggressive and cold-blooded killers.

Eastin and Griffiths made some ingenious interpretations on the theories (e.g. General Aggression Model) that explained the associations between violent video games and aggression. First, presence or the feeling of being in the virtual world and feeling in the physical world at once, should increase feelings of aggression because you are inside the action. From a layman’s interpretation, an individual experiencing the highest level of presence can’t make the distinction between the real world and the virtual world. Another factor is the point of view of the action whether you are actually doing violent actions. So first-person shooters come to mind as players are playing “as” the character (e.g. Gordon Freeman), instead of “with” the characters (e.g. Solid Snake). However, current research suggests that the lack of familiarity with VR sets hampers presence. So IMO, increasing realism through the five senses does not necessarily increase presence if the individual is uncomfortable with his environment or detects technological artefacts in their senses, say a stray pixel or sound distortion.

One interesting remark is that VG studies assumed that all violent actions are of equal value in influence because the decision to aggress reinforces our aggression cognitive schema, making any aggressive thoughts, feelings and behaviours more accessible to consciousness. However, the authors argued that shooting guns does not enter our mind as easily as does punching, pushing and such. Well, a good IMO explanation is that we don’t use guns as innately and easily as we use our fists and feet, nor is real guns prevalent in our lives. So, the authors proposed that violent actions that are closer to our cognitive schema of aggression would elicit more aggression than something that is violent, but far off from individuals’ aggression knowledge structure.

A final point of interest is whether you’re fighiting against a human opponent or a computer opponent. This brings up competition effect, where a highly competitive environment would make individuals be more aggressive. Competitive sports bring to mind, so you get the idea.


Participants: 219 males. Age range 18 to 31 (M = 21.38, SD = 2.51), mostly white, most of them are at least second-year students, 95% said they played a video game in the past, and average playtime is 70 minutes.  This was done so because most gamers are men, most game characters are men, generally more competitive than woman and IMO an extension to a previous study (Eastin, 2006).


Presence: The 32-item Presence Questionnaire, items were rated on a 7-point scale.

Hostile expectation bias: this is used for measuring aggression. Basically, participants are to read three ambiguous stories and were asked how the main character in the stories would do or say (behaviours), think (cognitions), or feel (feelings). Now this measure is used in previous studies for violent VG effects of aggression.

Games used: Unreal Tournament: Game of the Year Edition (shooting condition), Knockout Kings 2002 (punching condition) and Gran Turismo 3: A-Spec (control condition). These games were selected because all of them are competitive, considered as actions games and are clearly distinct from each other.

Gaming environment: Two types of gaming environment: standard joystick using a PC. The virtual reality condition is done through a VR headset and game-specific joysticks. For UT, the joystick looked and felt like a gun. However, movement is controlled through the joystick so IMO there could be some discomfort in playing an FPS in a VR setting. As for the punching and control condition, they seem pretty good, wrist sensors for punching and game mate for movement, sterring wheel and pedals.


Participants are randomly assigned to one of six conditions (one of the three games and one of the two gaming environment). Participants were given a training session of a maximum period of 20 minutes, so they become familiar with it. Then at a later time (undefined), participants played two 10-minutes testing sessions and were told either they would play against a computer or human opponent. Interestingly, these testing times were used by another study that had “found” significant correlations between aggression and violent VG play (Anderson & Dill, 2000). But the time lag between the training session and testing session is worrisome as participants may forget their training session and become naive to the gaming environment. Another interesting procedure is that game characters would not “die” in a gaming match, so participants are playing without interruption from losing or winning.


Results were conducted through ANOVAs.

Presence: participants in the console condition reported higher presence scores than participants in the VR condition. Those in the shooting condition reported higher presence scores than the punching condition.

Hostile expectations bias: Overall participants in the punching condition reported higher scores for hostile expectations than the shooting or control participants, whereas there were no significant differences between the shooting and control condition. Between consoles and VR sets, there was an overall significant difference in that the VR condition had a higher score than the console score. But when looking between games in each conditon, they found that there were no significant differences between the games in the console condition, but there were significant differences found in the VR condition: Again, they found that the punching condition was significantly higher than the shooting and control condition. But there were no differences between shooting and control condition. The authors made further analyses by looking into the individual component of the hostile expectations bias measure.

Aggressive behaviours: Overall, punching condition significantly higher than shooting and control. Shooting and control are not different from each other. For gaming environment, no differences. In the interaction between game and environment, a difference was found that those playing in the VR punching game had a higher score.

Aggressive thoughts: Overall, punching condition significantly higher than control. Shooting condition is not different from punching or control condition. For gaming environment, no differences. In the interaction between game and environment, a difference was found that those playing in the VR punching game had a higher score.

Aggressive feelings: Overall, punching condition significantly higher than control. Shooting condition is not different from punching or control condition. For gaming environment, no differences. In the interaction between game and environment, there were no differences.

Human vs. computer opponent: No significant results found.


Do the results mean that concerns over the negative effects of violent virtual games overblown. Not quite, there are some interesting factors to consider about the results of the study.

First, Eastin and Griffiths mentioned “functional interactivity”, that is the level of comfort and coordination between controller and visual feedback, might have influenced the level of presence and the association with aggression. As mentioned earlier, movement controls for an FPS game in a VR setting looked pretty awkward to me, even if participants were given prior training. I think it’s not enough to train them once and have them tested some weeks later. Nevertheless, there are some studies that have found differences in aggression in varying one or two sensory realism. Specifically, Barlett et al. (2008) study’s used a rail-game’s light gun vs. a standard controller, minimizing any discomfort and removing the need for movement. A similar setting occurred for the Persky and Blascovich (2006) study where participants were only required to shoot, either in a VR setting or in a standard gaming environment.

Second, in line with functional interactivity, IMO, I believe lower levels of sensory realism should tell us something. Instead of using VR sets that haven’t entered popular use, we should test for visual realism that is a large-scale version of what’s popularly used. That is using projectors and surround sound system in a room, something that is used in police training. This kind of environment is within reach for gamers, but is not adopted by any game developers, so this is a potential research venue.

The authors made the games so that participants are constantly in play, which is a sort of god mode. They argued that this study would be different from a natural game setting. However, IMO this is just an assumption, a lot or not a lot could happen in what… 10 minutes? The reasons weren’t discussed, but it partially answered a question I had about using Serious Sam’s crazy waves of enemies. The potential use of God mode in a research setting is that we could control individuals’ self-efficacy in a game, and as Eastin and Griffiths mentioned motivations and expectactions, which could have influenced presence and aggression. IMO, sometimes we experience a loose sense of time, called flow experience, when we are enjoying ourselves because it’s perfectly challenging. Perhaps flow has an influence over our aggression schema. On the other hand, god mode allows us to investigate whether a constant barrage of violent acts in a video game really does increase access to our aggression cognitive schema.  I am encouraging that using god mode is a great tool for experimenters.

A theoretical consideration from my reading of this study is whether media violence does not really relate to real-life aggression. What if interactive media violence relates to something completely new, what if we are really aggressive virtually? Virtual aggression…acting aggressively upon provocation through a mediated environment. The effect of violent VG might be strongest in virtual aggression, but much weaker for the general concept of aggression because of the lack of behavioural and cognitive connection. As the study has shown that the VR punching condition elicited higher hostile expectation scores.

The authors included other factors to consider: characters’ gender in relation to player gender, developmental consideration of adolescents and children, naturalistic gaming sessions and making more definite conclusions about presence in relation to aggression.

Finally, if we take to heart the result of this study, it conjured something in my mind that potentially trumped many anti-video games advocates on the subject of realism and interactivity. But, I’ll let a youtube video explain for me: The uncanny valley.

Eastin, M. S., & Griffiths, R. P. (2006). Beyond the shooter game: Examining presence and hostile outcomes among male game players. Communication Research, 33(6), 448-466.


One thought on “Differences in video game interface, content and context in relation to presence and hostility (Eastin & Griffiths, 2006)

  1. Pingback: Technological advancement and violence exposure Level 2 (Barlett et al., 2008) « VG Researcher - Psychology

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