Inside Out: The Case for Virtual Doppelgangers and How to Prevent Improper Use

The research is in, and manipulating the actions of digital avatars (any digital character a user can control) that look like a user (virtual doppelganger) can affect the user’s behavior. Some programs that contain virtual doppelgangers and make users behave better already exist, like software that ages your virtual doppelganger so you will save more for your retirement, while programs that would instigate negative and antisocial behavior have already been imagined, like software that makes you watch your virtual doppelganger torture people. Decreasing the effects of harmful uses of virtual doppelgangers and other digital avatars might be as simple as acknowledging the power of that manipulation.

Daryl Bem’s Self-perception theory posits that a person’s attitudes are informed by their actions. As counter-intuitive as this theory may seem, researchers have used behavior to change the attitudes of subjects. In Leon Festinger’s 1957 Cognitive dissonance research, participants who were asked to do a task for for a low dollar amount were more likely to say they enjoyed that task over users who were paid well to complete the task. The less you were paid, the more you must have enjoyed the task to do it in the first place. The more you were paid, the less you had to like the task to want to do it. Mark Frank and Thomas Gilovich found that athletes in black uniforms performed more aggressively than athletes in white uniforms. If the only difference between an aggressive athlete and a calm athlete are the colors of their uniforms, the uniform must be altering the thoughts of the players.

If a change in a person’s actions can change their attitudes, can changing an avatar’s actions change how a player uses that avatar? Jorge Pena, Jeffrey Hancock, and Nicholas Merola found that users who controlled avatars dressed in black cloaks made their avatars participate in greater amounts of antisocial behavior than users who controlled avatars dressed in white cloaks. Even while controlling a body that is not their own, users are still feeling a digital version of the Self-perception theory, known as the Proteus effect. The Proteus effect states that user actions are altered by the features of their virtual avatar. Nick Yee and Jeremy Bailenson have found that users with taller avatars in virtual space were more confident negotiators, users with more attractive avatars were more social, and users with male avatars used more masculine gestures both in virtual space and immediately after leaving the virtual space. The effects of Self-perception theory and the Proteus effects apply to user controlled avatars, altering attitudes based on external avatar features.

If the appearance of a virtual avatar can affect how a user manipulates that virtual avatar, can the independent actions of a virtual avatar affect the actions of a user? Jesse Fox and Jeremy Bailenson found that users who watched their virtual doppelgangers exercise, exercised more in the 24 hours following the virtual doppelganger observation. Sun Joo Ahn and Jeremy Bailenson also found that advertisements which were altered to show participants of their study consuming soda from a particular brand led participants to have better associations with that soda brand. Not only do the actions of your virtual doppelganger influence your actions, but the displayed preferences of your virtual doppelganger can influence your preferences.

Virtual and augmented realities are unique in their ability to both control all aspects of an environment and to require very specific actions from users. Requiring a user to exercise, smile, or wear a white robe might positively affect users in a way that other media has not been able to; however, since this technology is amoral, creators can easily require a player to act in ways that are physically harmful, depress moods, and partake in antisocial behavior. A more cynical creator might take advantage of how users are affected by their avatars and create an experience which intentionally harms users at the benefit of the content creator.

As a distributor of content, I don’t want to discourage anyone from making any type of content, regardless of its impact on the user. I love introducing people to SUPERHOT VR, even though they are often made uncomfortable when they are told to shoot themselves in the head. It is our responsibility as stewards of the AR/VR community to self regulate by educating users, similar to the video game industry. Knowing that content is advertising a product highlights the intentions of the creator. To expose a creators intentions in their manipulation of a virtual avatar is to provide agency to a user. Do you, the user, want your behavior to change?

References

1. Ahn, S.J., & Bailenson, J.N. (2011). Self-endorsing versus other-endorsing in virtual environments: The effect on brand attitude and purchase intention. Journal of Advertising (pp. 93–106). Retrieved from https://vhil.stanford.edu/mm/2011/ahn-ja-brand-attitude.pdf
2. Bem, D.J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (pp. 1–62). New York: Academic Press. Retrieved from https://www.researchgate.net/publication/277682193_Self-Perception_Theory
3. Festinger, L. (1957). A Theory of Cognitive Dissonance. California: Stanford University Press.
4. Fox, J., Bailenson, J. (2009). Virtual Self-Modeling: The Effects of Vicarious Reinforcement and Identification on Exercise Behaviors. Media Psychology (pp. 1–25). California: Stanford University Press. Retrieved from https://vhil.stanford.edu/mm/2009/fox-mp-selfmodeling.pdf
5. Frank, M.G. and Gilovich, T. (1988). The Dark Side of Self- and Social Perception: Black Uniforms and Aggression in Professional Sports. Journal of Personality and Social Psychology (pp. 74–85). New York: Cornell Press. Retrieved from http://www2.psych.ubc.ca/~schaller/Psyc591Readings/FrankGilovich1988.pdf
6. Peña, J., Hancock, J.T., Merola, N. (2009) The Priming Effects of Avatars in Virtual Settings. Communication Research (pp. 1–19). Retrieved from http://journals.sagepub.com/doi/abs/10.1177/0093650209346802
7. Yee, N., Bailenson, J. (2010). The Difference Between Being and Seeing: The Relative Contribution of Self-Perception and Priming to Behavioral Changes via Digital Self Representation. Media Psychology (pp. 195–209). California: Stanford University Press. Retrieved from https://vhil.stanford.edu/mm/2009/yee-mp-perception-priming.pdf