Transformed Social Interaction: Exploring the Digital Plasticity of Avatars

 Bailenson and Beall begin with two questions that get straight to the heart of the matter, "What does it mean to be you? How drastically can a person change and still remain...the same person?"  These bring to mind the work of author Terry Pratchett.  In one of his books (The 5th Elephant) a king explains about his father's axe.  Sharp as could be at one point, in time the blade became rusty and needed to be replaced.  Awhile later the handle split and was replaced.  Yet the object remained wholly his father's axe.  Though we may alter pieces of our personality, our core remains.  Never before has it been so easy to alter ourselves and 'try on' new identities than it is with the advent of collaborative virtual environments (CVE) and the avatar.

The authors refer to this ability as Transformed Social Interaction (TSI).  They divide TSI into three dimensions.

1. Sensory Abilities: Humans can acquire augmented perceptual abilities in digital environments.  These could be algorithms that provide real-time summary statistics about the movement and attention of others or the avatars of others that only you can see.
2. Situational Context: Each individual can adjust the geography of the room to suit his or her needs.  Imagine a classroom in which every students is at the front of the class with his or her peers behind.  Furthermore the ability to use 'pause' and 'rewind' features while an avatar is set to 'auto-pilot' greatly alter the ability to comprehend and be productive during interactions.
3. Self-Representation: As we have seen in other articles, the ability to alter the way in which one's avatar is rendered can have an incredible impact on online interactions.

It is on this third dimension that the chapter focuses.  They outline a number of ways that transforming one's self-representation can have drastic effects on online interaction.  The first of which is altering one's appearance based on markers captured from an audience member.  This will promote implicit feelings of similarity - which has been demonstrated to effect social influence. They continue with the concept of 'Team Face'.  Aspects of each team member's physical appearance are incorporated into an avatar that reflects all of them. Once again, this similarity has large effects on our perceptions of the interaction.

This science transfers nicely from the visual to the acoustic.  Voice can be transformed in various ways and for numerous ends.  The classic example of  using a voice changer to disguise one's voice is just one aspect of what the technology is capable of doing.  The author's believe that the effects of altering one's voice would be similar to that of altering one's physical appearance.

Bailenson & Beall continue by exploring the effects of digital chameleons - see an earlier post for more on how altering one's behaviour to appear similar to another person can alter the interaction between the two.

The author's finish by concluding that these dimensions make it difficult to trust online interactions.  What is real and what is virtual?  TSI offers users a vast amount of options when it comes to altering online interactions.  Those who fully understand the impacts of this are positioned to exploit those who are unaware.  with everything going online it will be important to be critical when accepting what one discovers online.  Even that political debate could be biased due to TSI.

Bailenson, J. N., & Beall, A. C. (2006). Transformed social interaction: exploring the digital plasticity of avatars. In R. Schroeder (Ed.), Avatars at work and play (pp. 1 - 16). Netherlands: Springer.

Body and mind: a study of avatar personalization in three virtual worlds

With millions of users involved in virtual worlds, the authors of this study have decided to focus on the avatar as a visual representation of identity.  Beyond the sophisticated ability to control body language through the virtual representation, this took chose to focus on the physical form of the avatar itself.  They cite references that document the use of avatars to represent aspects of one's 'ideal self'.  While the whole may not be representative of the user, they aid in the establishment of self-esteem.

The researchers noted that, "Most participants (68%) create avatars looking different from themselves..." Female users are more likely to create avatars that are idealized versions of themselves, whereas male users tend to create avatars that stand out more in the virtual world.  Age-wise, older users were more likely to create an avatar based on an ideal self.  When avatars differ from the physical self it is usually in the areas of attractiveness, physical fitness, and standing out from the crowd (in that order).

Taking it to the next logical step, the researchers explored the attachment participants had to their avatars.  Those that created an idealized self were more attached to the avatar, with older participants once again having a larger level of attachment to this ideal self.

When exploring the psychology of the avatar versus that of the physical user the researchers were able to replicate earlier studies point for point.  Avatars are more conscientious, extroverted, and less 'neurotic'.  they also found avatars to be less open than the physical self.  Those participants that reported the smallest differences between their physical and virtual identities ranked themselves as being more pleased with their avatars.  Furthermore, the more time a user spends online, the less difference will be found between their online and offline identities.  The disconnect between who they are offline and who they are trying to be online is simply too much to handle.

It is true that we enjoy altering ourselves online to experiment, but we often abandon those experiments when they are too far from our 'real' selves.  It would be interesting to explore case studies of individuals who create avatars who are very different from their physical selves in both appearance and personality.  Would the dissonance become overwhelming in short order? I also wonder who those with dissociative disorders would react in a similar study.

Ducheneaut, N., Wen, M., Yee, N., & Wadley, G. (2009). Body and mind: a study of avatar personalization in three virtual worlds. Conference on Human Factors in Computing Systems: Proceedings of the 27th international conference on Human factors in computing systems, 1, Retrieved from http://portal.acm.org/citation.cfm?doid=1518701.1518877 doi: 1518701.1518877

Digital Chameleons

The 'Chameleon Effect' refers to the, "Tendency for mimickers to gain social influence in computer-mediated communication."  In past studies, Bailenson & Yee report, participants show a preference toward confederates that mimick non-verbal communication and gestures.  They rated these chameleons as more likable and their interactions as more smooth.  The authors argue that, "Both unintentional (automatic) and intentional mimicry facilitate and express social affiliation and prosocial behaviour..."

For the study Bailenson & Yee used a Collaborative Virtual Environment (CVE) where participants in different locations had their verbal and nonverbal behaviours mapped onto online avatars.  Because the CVEs are completely digital it is easy for a user, through the use of an algorithm, to mimic other users, even to go so far as to have one's avatar appear differently to individual users.

The findings were interesting.  Bailenson & Yee note that participants viewed the mimicking agent as more persuasive and likable, and were less likely to turn their heads such that the chameleon was outside their field of vision.  They also state that the participants no the whole did not explicitly notice they were being mimicked.

The authors do note that the mimicking agent acted in a way that was contingent on the participants behaviour, and that this could have confounded the study as it was an uncontrolled variable. 

Bailenson & Yee continue by telling us that the chameleon effect continues, even when the participant knew the agent was the embodiment of a non-human artificial intelligence. 

This suggests that with a few relatively simple lines of code a person can create an online avatar that will be positioned to create the best possible impression in a virtual environment.  It will be interesting to discover what effect this could have in online commercial dealings or business relationships.

Bailenson, J., & Yee, N. (2005). Digital chameleons automatic assimilation of nonverbal gestures in immersive virtual environments. Psychological Science, 16(10), 814-819.

Identity Management in Cyberspace

 Suler identifies five interlocking factors that allow us to navigate how we manage our online identity. 

Level of Dissociation and Integration
Our personalities are created of many different facets - roles we play in different situations.  Online we are able to parse these aspects according to the specific environment.  In each situation we make a decision as to how much we share about ourselves.  While this occurs in the offline world as well, it is more concentrated online as many of the groups we join reflect only one small aspect of our personality.  Suler states, "The desire to remain anonymous reflects the need to eliminate those critical features of your identity that you do not want to display in that particular environment or group." 


Positive and Negative Valence
Our different components can each be assigned a value of positive or negative.  Showing empathy would be a positive trait, while criticizing the flaws of others unduly would reflect negatively upon our character.  Beyond the basic altruistic values of good and bad, we can feel negative emotions towards aspects of our personalities that are seen as positive by others.  All of this charges our online interactions.  Suler suggests that those who behave in a negative fashion online are discharging a negatively charged aspect of their personality - venting their emotions, or trying to work through that aspect of their personality in a 'safe' way. 

Level of Fantasy or Reality
The situation itself dictates rules for how one presents him- or herself.  In these situations one does not pretend to be something other than what they are - well, not often anyway.  In other situations, like a MMORPG, it is expected that a person will give themselves over to character.  Some aspects may change, physical appearance, occupation, name, while others may stay the same.  Suler states, "The tricky phenomonological issue with the real versus the fantasy self is this: what is one's TRUE identity?" 

Level of Conscious Awareness and Control
We are not always aware of how we present ourselves.  When we engage in online interactions it is possible that aspects of our personality we try to keep hidden will leak out.  The name we select on a whim may reflect a deeper symbolic meaning to which we have not paid attention.  It is possible in this way for online characters to 'take on a life of their own'.  Sometimes this allows us to gain insights into personalities.  Yet others may resist looking more closely at themselves in this way and continue to, "Live under the illusion that they are in control of themselves."

The Media Chosen
We express who we are in a myriad of different ways - from the clothes we wear, music we listen to, movies and television we watch, and the other media we consume.  " Extending the logic of the statement, 'The medium is the message,' we might even be so bold as to say, 'The medium is the self.'"  The very form we chose to express our identity online says much about who we are.  From linguistic- and semantically focused text messagers to visualizers who prefer detailed avatars, what we say is not always as important as how we say it. 

Suler has identified these five traits, but fails to explain the implications of these aspects.  Are these sliding scales that could be used to create a typography of online personalities?  It is unfortunate that Suler ends the article at this point, without further exploring the issues raised.

Suler, John R. (2002). Identity management in cyberspace. Journal of Applied Psychoanalytic Studies, 4 (4), 455-459.