Okay this is my last one for today… I hope. Now following Dr.Eastin’s academic trail, he wrote an article published last year. Since I’m trained in APA, I do find that the words in the title seemed so wrong. All capitalized, how strange…
Adding depth and breadth to the general aggression model, this paper presents three experiments that test the relationships among user and opponent gender representation, opponent type, presence, and aggressive thoughts from violent video game play. Studies 1 and 2 suggest that females experience greater presence and more aggressive thoughts from game play when a gender match between self and game character exists. Studies 2 and 3 indicate that playing against a human opponent (rather than a computer) increases aggressive thoughts. Finally, although Studies 1, 2, and 3 indicate that playing as a female against a male opponent increases aggressive thoughts, Studies 1 and 2 suggest that playing as a male against a female opponent consistently and significantly decreases aggressive thoughts.
In this study, Eastin examined the relations on the effect of gender representation in a series of experiments. Now, gender representation means the player’s character’s gender and the opponent’s gender, either from the character or from the opponent player. Other examined relations are the concept of presence, agency of the opponent (human player vs. computer player).
Now, the dependent variable is aggressive thoughts of players (cognition).
One of the aggression factors of violent video games is identification, it is especially true for FPS games. Some games allow a creation process of which players can represent themselves within the game world, such as height, weight, clothes, skin color, hair style and gender. Hence, this creates a connection and an emotional investment between one’s identity and the avatar’s identity. As these identities are connected, in-game behaviours activate cognitive schema within individuals. In FPS games, psychologists would usually view these in-game behaviours as aggressive and violent for good reasons. Thus, repetition of aggressive behaviours and a strong identification to violent characters would lead to more easily accessible cognitive schema of aggression. Mind you, this does not translate directly into aggressive behaviours or attitudes as said in other studies.
Eastin wondered whether gender would be a factor in character identification. For example, a women who plays a female game character would be more likely to aggress. IMO, that depends on how someone views femininity, since it being female can be as caring, compassionate and non-violent. Well I’ll have to think on it further…
Tied to gender representation is presence. I’m a little weak on the concept of presence, as it is a relatively recent concept in social science. Eastin follows presence as “feeling in one even when located in another.” (p.354) However, there are three aspects of “presence”: social, spatial and self. What Eastin is more likely looking at is self-presence and social presence, which fits in this study’s operational definition.
In relation of presence to FPS games, players experience a strong sense of presence, since it is first-person view, we can interact with game objects, or get up-close and personal with the enemy, but importantly serve a point of reference within the virtual environment. I’m not sure if you can follow my summary, but that’s the best I can do.
Three experiments are performed, each experiment have independent samples and slightly different procedures.
General study sample
It seems the sample consisted of only female players. I guess including male players would complicate the experimental design and statistical analysis. I should ask Dr.Eastin about that. Average sample size is about 75 participants, largely recruited from intro communication classes. Average age is about 20 years, an average of 75% are Caucasian and average daily video game time of 30 minutes. However, this does not tell whether the games they played were casual or hard-core. From what I know, there aren’t many gamer girls.
General experimental procedure
Game used was Unreal Tournament: Game of the Year Edition. Two maps are used in this study, a training map and the testing map. The testing map was created for this study, it consisted of a central mirrored column to reflect the players’ avatar’s gender as a mean to remind them of their virtual gender representation. This method of using a mirror is validated in other studies, IMO other methods I could include is gender-typical sounds, height, and showing more of the arms. Game difficulty was set at intermediate. The same avatars are used throughout the study. Weapons were restricted to only three types. Game opponents are computer-controlled.
Participants go through two sessions. One 20-minute maximum training session. (no bots, no mirrors, no virtual gender representation and map is different from testing). Participants are randomly assigned to one of testing variables. The testing session lasted 20 minutes. Questionnaires were given before the training and after the testing session.
General experimental measures
Presence questionnaire (PQ) contained 32 questions on a 7-point semantic differential scale. (no clue what it means) These questions are categorized into 4: control, sensory, realism and distraction. Control refer to the responsiveness of the environment, sensory refers to engagement and involvement, realism refers to level of connectedness to sensory information and distraction refers to levels of perceptions of control. (p.358)
Aggressive thoughts (cognition) are assessed using the word completion task. Basically, participants fill out words that have a missing letter. So participants fill out h_t either as “hit” or “hat” (coded as aggressive and non-aggressive respectively). Participants are given 5 minutes to fill out a maximum of 98 words.
Experiment 1 examined the relations of players’ gender and matching or mismatching avatar gender to presence and aggressive thoughts. It is hypothesized that a matching gender (i.e. female player with female avatar and female avatar opponent) would increase presence and aggressive thoughts.
Players are told that the opponents are computer-controlled.
Testing variables: player avatar gender: male vs. female
Opponent avatar gender: male vs. female
Power analysis was used to analyze results. What is G*Power anyway?
The results showed that matching gender was found to increase presence, but there were no significance results for opponent gender.
Interactions effects were found for aggressive thoughts. Matching gender elicited more aggressive thoughts. A higher effect was found when the opponent is male. On the other hand, male avatar against a female opponent elicited the least amount of aggressive thoughts. Male avatar vs. male opponent is slightly less than the matching gender effect.
IMO, I guess our real life experience and social values to interact differently to either gender has a hand in it. In addition, it seems that gender role-playing seems to have a hand in it too, since female players have lower aggression scores when they played as a male against a female opponent, perhaps women thought that men should be acting more gentlemanly towards women. Alternatively, the act of hurting another women as a man elicited more empathy rather than aggression.
Easting wrote something that confuses me: “Why did the male avatar representation increase hostility toward the female agent opponent?” this question seemed to contradict what the data results showed earlier. Maybe he’s saying that the aggression measures is influenced by other factors…
This experiment examined the relation on whether it is a player or a computer that is controlling the opponent to presence and aggressive thoughts. Much of the rationale is based on social presence and competition research. An interesting note about competition is that opponents who are at arms’ reach are less aggressive than those playing against a computer.
Participants were randomly assigned to and told they would play against either a human or a computer. Participants assigned to the human opponent condition would be playing alongside a female confederate to ensure that they are thinking that they are playing against a human opponent. But all opponents are computer-controlled. Participants are randomly assigned to avatar gender and opponent gender.
Please note that the analysis for human vs. computer would be done with only female avatars and female opponents.
Using the same power analysis, results showed there were no significant differences between human vs. computer opponent on presence. However, there are significant results showing that human opponents elicited more aggressive thoughts than computer opponents. Interaction effects revealed that matching gender elicited high levels of aggressive thoughts regardless if it is a human or computer opponent. Even playing as a male avatar seems to elicit the same levels against a human player, but aggressive thoughts were lowest when playing against a female computer opponent.
IMO, it seems that when we play against human opponents, we play against the players who’s controlling the opponent and not at the in-game opponent. It seems right because I’m thinking that we behave and think differently against who or what we play against. In addition, I think we feel more accomplished or challenged when we play against someone else.
Eastin seemed to argue that while matching avatar gender has increase effects on presence and aggressive thoughts, opponent characteristics are more complex.
This experiment would focus on opponent gender effects on presence and aggressive thoughts.
Experiment specific procedure
Participants would only play the female avatar.
Testing variable: opponent type: human vs. computer
Opponent gender (in-game and player together): male vs. female
One caveats in this design is to examine the effect of human players playing in opposite-gender avatars.
Opponent gender effects on presence have found no significance. As for aggressive thoughts, male opponents elicited more aggressive thoughts than female opponents.
Eastin concluded that having strong character/avatar identification increase our sense of presence and aggressive thoughts. However, presence seems to be independent of aggression, IMO the connection between presence and aggressive thoughts have not been examined. Opponents’ gender has a role in players’ level of aggressive cognition in that fighting a male (human or computer) opponent has higher effect than female opponents. But in a twist, playing the opposite gender seemed to have complex effects in that female opponents elicited the least amount of aggressive thoughts. Another factor that was mentioned and examined briefly across the experiments was arousal, it was found not to be a significant factor.
IMO, what to I get out of this study is that our aggressive thoughts is dependent on how we are represented in the virtual world, whom we interact with in a virtual, violent and competitive environment.
Eastin stated some study limitations, first is that there were no analysis on in-game behaviours. So we don’t know whether in-game aggressive behaviours might elicit more aggressive thoughts. Age is an issue on how youth develop aggressive scripts (don’t know what’s he getting at) Perceived competition is not measured, so we don’t know whether players see the game as a competition or something else. In any case, Eastin did another study that addressed some of these concerns.
IMO, other limitations include that this only applied to female players. There are gender differences between female and male in how they think and behave, one that comes to mind is that females are probably more attentive to social cues and gender cues in video game environments, so are more likely behaving in-character, whereas guys are probably more concentrated on the competitive or instrumental aspect of the game (i.e. win, win, win). Hence, I want a replication study for male players. Second, game gender representation (avatar and opponent) might have an influence, that is I wonder if normal-looking women might elicit different reactions than hyper-sexualized women (i.e. huge breasts and skimpy clothing). Third, I wonder if realism and conversely presence has nothing to do with, perhaps a very old FPS game might show that opponent characteristics have a larger role. Fourth, the level of identification and presence might be a moderating factor because other studies have such effects. Fifth, what about gender differences in gaming performance, I mean I would believe women are more aggressive that a male player scored a frag at them, partly of the alleged stereotype they form about guys and video games or perhaps remember more about being fragged by a guy than by a girl. Sixth, what about game experience? I mean do veteran players tend to ignore gender and focus on more performance related aspects, like bot difficulty. Seventh, what about gender-neutral characters, animal characters or cute characters, I’m sure they would elicit different levels of aggressive thoughts.
Eastin, M. S. (2006). Video game violence and the female game player: Self-and opponent gender effects on presence and aggressive thoughts. Human Communication Research, 32(3), 351-372.