In their study on the “Proteus Effect” Yee and Bailenson (2007) discuss past research on virtual environments and the capacity of users to easily manipulate their “digital self-representations.” The main question Yee and Bailenson propose is whether the avatars we create (which they define as the “primary identity cue in online environments” (p. 274)) can influence our behavior in the virtual, and perhaps real, world. The define the ‘Proteus Effect’ as how “…an individuals behavior conforms to their digital self-presentation independent of how others perceive them..” (p. 271). They also point to past research on how people match their behavior to uniforms they are assigned (i.e. more aggressive if assigned black uniform). The authors emphasized how important it was for their study to prevent Behavioral Conformation (we act more like how others perceive us) from interfering in their results, and I think they did a good job excluding it by keeping confederates blind to the attractiveness/tallness of participants. Yee and Bailenson found a significant correlation between the physical features of participants avatars and their behavior. However, I question how well their findings apply in real online interactions, where behavioral conformation and other factors complicate identity and behavior.
Since I encountered problems with Second Life, and don’t subscribe to any online video games, I decided to question my boyfriend Dave (who has much more experience donning an avatar in online spaces) about his experience playing World of Warcraft. World of Warcraft is perhaps the most well known Massive Multiplayer Online Role Playing Game. The environment is similar to the ‘real world’ in that it has trees, water, cities, roads etc. However, WoW is clearly based in a fantasy world in which the “
When creating a character for WoW, you have the option of first choosing a ‘side’ (Hoard vs. Alliance) and then choosing among the different races within each category. You can then modify your character’s features, including skin color, hair, clothing (that you find later) etc. Dave’s avatar was of the “Undead Warlock” class, part of “the hoard” which is considered the ‘evil’ side in the war. He described his character as about average human height but on shorter side, of all the classes. He was not particularly attractive—skinny with wrinkly dark skin, slouching posture, and dark purple hair. The main influence of Dave’s behavior that he noticed was that being of a race designated as ‘spell caster’ meant and during group raids his character is supposed to stand back and cast spells while others like warriors/rogues are in the front physically damaging enemies. It seems to be a rule of the game that you should conform your behavior to the role defined by your race. This was one type of effect on Dave’s behavior due to his avatar that seems to support the findings of this study, since he was conforming to other players expectations of his avatar, as Yee and Bailenson argue, “…in the same way that subjects in black uniforms conform to a more aggressive identity, users in online environments may conform to the expectations and stereotypes of the identity of their avatars.” (p. 274).
While he tried out a character on the
Dave’s characterization of the behavior of people with each type of avatar seems to reveal that those playing as more attractive characters disclosed more about their feelings and personal reactions to events in the game, while those who were less attractive, seemed to stay in role as a sort of ‘monster’ who probably wouldn’t be as social as a human. This could provide some support for Yee and Bailenson’s conception that, “in line with self-perception theory, [players] conform to the behavior that they believe others would expect them to have. We term this the Proteus Effect.” (p. 274). Another interpretation I considered is that those who take themselves more seriously, and want to produce the best self-presentation will probably be more likely to choose one of the attractive avatars on the
Dave was highly resistant to the idea of his behavior being influence by the physical characteristics of his avatar. However, he did say that he does not choose characters he can’t identify with—implying that he does identify with his avatar on some level. For example, he didn’t choose to be a gnome since felt he couldn’t identify with this race, which typically has a “whiny voice” and a cutesy, perhaps wimpy appearance. This may also play into his desire to have an avatar that presents himself as ‘tough’ or a worthy addition to a team in the game.
As far as explaining why Dave’s behavior wasn’t ( or he feels it wasn’t) affected by his choice of avatar, maybe a RPG can’t be compared to a Second Life type scenario where people are simply interacting as generic humans. Since in World of Warcraft and other MMORPG’s, players are taking the role of a specific character type (that comes with alliances, specific skills and goals) they tend to aim for those goals, but don’t feel like they are the character as much as those interacting as an avatar more similar to themselves, such as in Second Life.
Near the conclusion of their paper, Yee and Bailenson argue, “The set of studies presented in this paper makes clear that our self representations have a significant and instantaneous impact on our behavior.” While they have made some interesting arguments I think it would be taking their findings too far if we assume we can extend them to more realistic virtual spaces where behavioral confirmation as well as reduced identification with an avatar can interfere. Yee and Bailenson did point out that it would be interesting to “…pit the Proteus Effect against behavioral confirmation. In other words, future work should examine an experimental paradigm in which participants believe that they are attractive, whereas other interactants perceive them as unattractive.” (Pg 286)
Comments:
http://comm245brown.blogspot.com/2007/11/10-looks-always-matter.html
https://www.blogger.com/comment.g?blogID=5400576841210402935&postID=7168728462796344688
2 comments:
Hey Amber,
Thanks for commenting on my post. Do I think the higher level contributed to other peoples’ willingness to interact with me? Yes. This was actually the first thing that came to mind when I started thinking about the assignment. I don’t know why I forgot to mention it… In an MMORPG, the main goal of players is to gain skills, complete quest, etc. Thus, better items and higher levels are more attractive than actual looks. I could probably explain that a little more with theories (McKenna’s factors and conventional/assessment features…), but I don’t want to spam your post too much. You pointed out the something similar. People in WOW tend to concentrate on goals in the game, not necessarily interacting with others or relating to their characters. I really enjoyed your post, mainly because it was not about Second Life. The connections with theory were very well thought out. I liked how you quoted all the important parts of the paper. (It makes studying for the final really easy). You managed to apply them to an online space that was not to be played solely for social interaction.
Amber, I think it’s very good that you analyzed someone else’s behavior, rather than playing a game with the knowledge of what you’re studying. Obviously Dave wasn’t completely blind to the ideas behind the questions you asked, but in speaking of previous actions while he wasn’t conscious of his avatar’s effect on his behavior, I think you probably got less biased observations than the rest of us experienced in playing these games.
For my post I played a first-person shooter game and found that my actions were very much based on my character’s role on a team. I think this is consistent to what Dave said about his behavior being very task oriented and based on his class’s role in game.
Post a Comment