Reading: The Proteus Effect

As I delve deeper into my project, I’ve been thinking about how to develop a narrative that the player will find compelling. Moreover, since my game hopes to help learners practice perspective-taking, how can I develop avatars that learners can identify with? I’m interested in better understanding the effects of the relationship between a player and their avatar.

Nick Yee, Jeremy N. Bilenson, and Nicolas Ducheneaut (2009) collaborated to research the “Proteus effect,” the tendency for people and their behavior to be affected by their digital representations, also known as avatars. In the article, they start off by discussing what influences a person’s perception of an avatar. They suggest that “an avatar that mimics the user may be judged to be more likeable or persuasive” (Yee et al., 2009, p. 287). This is due to research in communication and social psychology where people mimic the behavior and speech patterns of others to elicit more favorable responses from them. They found that when participants interacted with a digital agent who mimicked their movements, participants were more likely to find the agent agreeable. They also discuss the similarity bias, where people will find other people who are similar to them, in terms of appearance or beliefs, as more “attractive” and “persuasive” (Yee et al., 2009, p. 287). They note the similarity bias can be triggered by arbitrary similarities such as sharing the same birthday.

Another interesting point the article brings up is how deindividuation increases “the impact of the identity cues given to the participant” (Yee et al., 2009, p. 292). When a person is anonymous, instead of losing their identity, they shift from “personal identity” to “social identity” (Yee et al., 2009, p. 292). I find this point salient for my game. It will be interesting to observe how player’s behavior changes when playing certain characters depending on 1) other character’s reaction to their character 2) whether or not they control the character as a disembodied avatar or if they see through the eyes of the character.

However, the main point of the article is to explore how the avatar’s appearance changes the behavior of the player, for the short-term and long-term (as in beyond the virtual environment). To explore this topic, they conducted two studies, one where they observed the performance of World of Warcraft players who had short versus tall avatars and another where they observed participants’ behavior with short versus tall avatars in negotiation. For the latter study, they observed both the participants’ negotiation in a virtual setting and in subsequent face-to-face interactions. The researchers hypothesize that the avatar’s appearance provides an altered self-representation which leads to “changes in a person’s behavior” (Yee et al., 2009, p. 291). Moreover, the person’s self-perception of their own appearance leads to changes in behavior. For example, in a study conducted by Frank and Gilovich (1988), players who work black uniforms “observe themselves as if from a third party to infer their expected attitudes and behavior” (Yee et al., 2009, p. 291). In their case, since others perceived the black uniforms as more aggressive, the players became more aggressive.

In the WoW study, they found that the avatar’s height was a significant predictor of the player’s performance in the game. Likewise, in the second study, they found that participant’s with taller avatars negotiated more aggressively in both the digital setting and in real-world face-to-face interactions. The researchers realized that in the second study, the change in behavior in the digital setting was due to the self-perception effect based on “observation of the avatars’ appearance” (Yee et al., 2009, p. 306). In the real-world interactions, the self-perception effect was based on the participants previous behavior in the digital world (because they negotiated aggressively in the virtual world, they perceived themselves as someone exhibits that type of behavior).

The article culminates in an exciting prospect, “On the level of the virtual community, these results imply that the design and implementation of avatars may have an effect on shaping the emergent social norms and interaction patterns.” (Yee et al., 2009, p. 308).

I think it would be really beneficial to test different configurations of avatar appearances for my game, especially in regards to physical attributes beard/no beard, attire, height/weight. I’m also interested in understanding how the movements (walking gait, posture) of the characters will affect player’s perceptions of them.


Yee, N., & Bailenson, J. (2007). The Proteus effect: The effect of transformed self-representation on behavior. Human communication research33(3), 271-290.

Leave a Reply

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

You are commenting using your 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