The concept of avatarial experiences modifying human behavior in some sort of reverse feedback loop is likely a necessary part of the requisite believability factor. That is, if we as real people are to be manipulated (or at least behavior modified) by our created avatars, we need to be able to suspend disbelief regarding their virtual identities in order to accept them and to succumb to their influence. I think that our disbelief can be mitigated through the use of technological advances in the areas of display and interaction.
Let me start by framing the counterpoint with an exaggerated example of abstraction: a game of strategy and warfare, a game where royalty commands armies with the mission of decimating the enemy. The game is ancient and known the world over; the game is chess.
The point that I’m making is that chess is very much like the games played on modern computer and console systems in terms of challenge, value of play, and so on. The substantial difference is that chess is completely a game of abstraction and intellect. The pieces are representative of the positions of power they occupy in the game, but they do not evoke any particular emotional response based on appearance, mannerisms, etc. As a result, there is little emotional connection with the pieces as other than chits in the negotiation of play. And this means that there is no (or very little) avatarial connection between player and piece.
Maia Garau (2006) asserts that mediated interaction between person and avatar in a virtual environment is improved by leveraging our ability to communicate non-verbally. Simply put, people want to see character expression, and the more highly defined the expression, the better able we are to connect and identify with the avatar as well as understand and empathize with it. So back to our Chess pieces: there are not non-verbal cues; no expressive eyes, no posture of fear or bravery, and so on. There is only abstraction of the piece’s role, and this is not compelling for us to develop a relationship with the piece.
The application of technology to entertainment has yielded ultra-realistic game characters that are able to communicate intent and action through visual and auditory cues that are natural to human beings. While this trend was a logical outgrowth of regular improvements in technology, there has been the unintended consequence of people becoming emotionally entangled with these onscreen characters–hence their classification as avatars rather than game pieces. Without thinking, we become involved with our game characters to some degree, and that degree appears to be increasing as the level of sophistication in avatar appearance increases. This is not a coincidence, and at least some preliminary research suggests why this is happening.
The thought that now strikes me is that there may not be any requirement for the physical form in order for humans to establish, build, and ultimately nourish relationships–at least beyond a certain point of physical development (i.e., for babies, toddlers, etc.). In other words, we value a form and an image that we can conceptualize and hold in our minds, even if we cannot hold that form in our arms. Could it be that we are hard-wired to actually value the “content of our character” above those physical aspects–race, gender, age–that lead to bias?
Garau, M. (2006). Selective fidelity: Investigating priorities for the creation of expressive avatars. Avatars at Work and Play, v. 34: 17-38.