My Avatar, My Self

By Nikki Massie

IDS-510-90

Fall 2013

Welcome!

If you are able, please click the below video to view and listen to my talk on this subject. If you are unable to view and/or listen to this video, I have provided the entire body of my research below with additional graphics.

If you have any questions about this presentation or have trouble viewing it, please email me at nmassie1@live.ndm.edu.

I hope you enjoy!

Introduction

I have to admit that prior to doing this presentation, I didn’t give much thought to avatars. It almost seems as if they have become, as Dietrich notes, a part of our “natural condition” (Dietrich, 211). I see them all the time. My daughters use avatars on their Wii and Xbox video game systems (as do I, on occasion). My youngest daughter likes to use a site called iDressup which assigns her an avatar that she can manipulate and dress in a myriad of designer clothing. I’ve even used a relatively new app on Facebook called Bitstrips, which allows you to design a comic-like avatar which can then interact with the avatars of your friends (which they generate) in comic strip scenarios.

In their foundational study called “The Proteus Effect” (which will be discussed in more length) Yee & Bailenson define an avatar as, “a perceptible digital representation

whose behaviors reflect those executed, typically in real time, by a specific

human being” (Bailenson & Blascovich, 2004, p. 65).

But if I think about it, avatars, while something of a natural state to me, are also a “ready-to-hand” tool. They are the way in which we can be present in a virtual environment. Without an avatar, one cannot participate in online environments. One could not interact with others. One would have no way to be visible in certain realms of the online world. Just like Nancy Mairs’ wheelchair (Dietrich, 219) was the only possible way to transport her body through the world, in many virtual, interactive environments an avatar is the only vehicle through which you can participate.

But even all of that didn’t seem to warrant any attention to avatars. What first caught my attention, to be honest, was when my 14-year old daughter joined Instagram. While avatars are not necessary to participate in Instagram, I noticed that she began participating in a relationship with herself that was defined by a sort of avatar.  She used an application that took a picture of her face and allowed her to manipulate her features. I began to notice that she would do so in very specific ways. She’d change the orientation of her eyes to make them look almost Asian. She applied make-up (she’s allowed to wear make-up in real life but elects not to). She’d apply color filters to her hair. And to me this representation of herself looked a little overdone, but that’s not the point. The point is that this was the self she was presenting to the world through this image. Insofar as our representations of our selves speak to our thoughts, beliefs and motivations, I think it’s important to pay attention to them. So I decided that it was time to pay more attention to avatars.

As it turns out, there is a great body of work related to avatars and self-perception. As I stated before Yee & Bailenson have studied what they call the “Proteus Effect,” which essentially says that “our virtual bodies can change how we interact with others in actual avatar-based online communities as well as in subsequent face-to-face interactions” (Yee et. al, 285).

To be completely honest, reading that, as a parent, is a bit frightening. From where we stand right now, my daughter’s self-representations are pretty innocent and I think I know all the virtual environments and societies to which she belongs, but as Knutzen and Kennedy (2012) point out, the virtual world is a place to which young people in particular are attracted because there are so many opportunities to “try on different versions of the self” (272). Given the transformative and ever-changing nature of adolescent personalities, while my daughter’s self-representations may be innocent today, they may not remain so. But the scariest part of that is that I don’t really have that much control over where she presents herself or in what form.

So how do we present ourselves through our avatars? And how do those representations both reflect and change us? Those are the questions I sought out to answer for myself.

What do we put into an avatar?

To answer the first question, I turned to the work of a Yee et. al (2011) and their study, "The Proteus Effect," which persuasively argued that attractiveness of avatars in game play (and thus in virtual environments) has significant impact on both the results of the game but also on the players real world interactions immediately following game play. A subsequent study by Knutzen and Kennedy (2012) sought to expand on that work by studying whether we apply the same systems of evaluating attractiveness to avatars as we do to real people and, further, if we use the value of attractiveness to determine our social relationships in the online world.

A female Nintendo Mii avatar. CREDIT: Raeleka
Used under a Creative Commons license.

The answer, their study suggests, is yes. A group of adults and children were first asked to rank a randomized group of adults on a numerical scale according to their perceived attractiveness. The group was then shown a randomized group of Mii avatars (generated by the Nintendo Wii system) and asked to rate those avatars on the same numerical scale according to perceived attractiveness. The groups rated the Miis in a similar way to real people.

The group was then told they were to choose a Mii to be their partner in a doubles game. Members of both the adult and child group were more likely to choose a Mii they had originally rated as attractive.

To me what’s interesting about this study is the fact that they were rating a Mii. Now I’m not sure if any of you have seen a Mii before but it is by no means the most sophisticated representation of the human form. It’s actually rather cartoonish. So for adults and children to rate a Mii, they probably had to use some recall of features that they know from real people and apply them to the cartoonish figures. This recall in perception is instantaneous and subconscious. So we don’t know we are doing it but we do. It’s a reflex almost.

So the groups recalled their notions (whatever they may be) of what is attractive and unattractive and applied it to the Miis. For some blondes might have taken precedence, for others brunettes. For some a certain eye shape or fullness of the mouth might have seemed appealing for reasons we can never know. But the important part is that the participants made a seemingly “snap judgment” on the Miis and then — here’s the important part — used those snap judgments to decide with whom they should align themselves.

In her interview, Lisa Nakamura aptly points out that we carry over social norms and cues from one realm of society to others. One might think that because the virtual world is not corporeal and because an avatar like a Mii is so different from the nuances of an actual human face, that people might incorporate a different method of assessing value. It seems in this study that this is not so.

So what are we bringing to our avatars? The answer would seem to be that we are bringing our personal biases, preferences and hopes. But there’s more.

According to Thomas and Johanssen (2012), women in particular can also bring our iconic ideals to avatars — that is, we not only bring what we are but who we want to be. In their study, they asked a group of participants to make an avatar that best represented their real and true physical selves. They were then allowed to make avatars that looked any way the participant desired. Not surprisingly, when participants were unrestricted by the parameter of making an avatar in their own likeness, they created an avatar that looked quite different from themselves. In many cases avatars had stereotypically “desirable” features.  

So really in creating an avatar, we are not only making a virtual representation of self but also a values statement. Both men and women tend to make values statements concerning the ideal body. But many people make a value statement by not choosing a human at all, but an animal or a fictitious creature, choosing to abstain from any statements about the realistic human form.

How do avatars reflect and effect us?

Even then, however, studies have shown us that characteristics of avatars can have a certain effect on us. So our representation, as told through the appearance of an avatar, can both reflect us (especially with regards to our values systems) and affect us through our behavior.

To learn more about that, I return to the “Proteus Effect” research, which included two studies. For each study, participants were assigned either an attractive, unattractive, short or tall avatar. They were allowed to interact in a virtual game environment with those avatars and socialize with study confederates. Afterward their game performance was assessed along with their subsequent real world interaction.

The study found that players with avatars rated more physically attractive, as well as those that were taller, performed significantly better than players with unattractive or shorter avatars. Further, they found that players with taller avatars displayed more assertiveness in interactions immediately following their participation in the virtual environment.

So at first it seemed as if the avatar has the potential to give a confidence boost that can help us in our interactions in the real world. However, there are studies that show darker implications, like those described by Bailenson and Tricase (2013), in their study of the embodiment of the sexualized self through avatars.

Bailenson & Tricase's study suggests that
sexy avatars like these increase
self-objectification
and belief in rape myths in women. 

The studied looked at how the use of hyper-sexualized avatars can affect how women think about themselves and each other. Women were given avatars that were rated by a separate group as either “not very sexy” or “very sexy.” They then interacted in a virtual environment with a male confederate. Afterward they were given a questionnaire. In the questionnaire, the women who were given hyper-sexualized avatars reported more thoughts about their bodies, behavior that suggests self-objectification. In addition, they also reported a belief in “rape myths.” A rape myth is a set of beliefs about a victim of rape, such as the belief that if a person dresses in a certain manner they are “asking for it.”

Let me pause here and ask if that scares anyone else as much as it scares me?

The study points out some of the implications of this. With online gaming becoming more popular, and because the avatar is the ready-to-hand tool through which we experience these games and because women in particular are subject to creating idyllic avatars, including highly sexualized ones, I think we can safely hypothesize that there is a sizable portion of women who would be impacted if the Proteus Effect were taken as fact.

Insofar as these same women are likely to interact with other women in important capacities, the inherent danger of the Proteus Effect is a bit concerning. Say, for instance, a woman is, in fact, sexually assaulted and her attacker was brought to trial. Say the jury contained women who actively engage in virtual environments with highly sexualized avatars. Say that engagement changes her perception of self objectification and her belief in rape myths. Is this woman now an impartial peer of the woman who was assaulted?

The news is not all bad!

Now I want to switch streams for a moment because it almost seems as if I am saying avatars are bad. So far I’ve said that we pour our biases, preferences and potentially unrealistic ideals into them. I’ve also said that those representations, in turn, affect the way we relate to people in both virtual and real world environments. That all sounds bad, right?

I’m a bit hesitant to say it is. In fact, one question I was left with after doing all this reading was this. Firstly, assuming we are submerged into virtual environments often enough to sustain the Proteus Effect (and I’m not entirely convinced that’s true), in what ways can we manipulate virtual environments to create positive behavioral change?

That research led me to a company aptly called, “Avatars for Behavioral Change," a beta project wherein they are creating avatars that help promote healthy lifestyles. So in the virtual world they are creating, avatars do things like participate in appropriate amounts of physical activity and choose healthy foods and manage their levels of stress.

Avatars for Behavioral Change
is seeking to use avatars to
promote healthy lifestyle
habits. 

Also, the Centers for Disease control has developed virtual games to make people more aware of managing chronic diseases like diabetes. In these games, characters (which technically pre-date our current conception of avatars) move through games with the unique considerations of a person with diabetes. At intervals they needed to check their blood sugar, make wise food choices, see medical professionals. Participation in this game actually improved the management of diabetes for the diabetics who played the game.

So in the end, I think there’s a lot of potential for both positive and negative representations of self with avatars.

Being a person who is interested in the obesity epidemic, the Avatars for Change effort excites me the most, but more than that I think there is great aspirational potential. By being able to express our idyllic selves, we can do several things. Firstly, we can visualize ourselves as those idyllic selves. Secondly, we can become acclimated to navigating social interactions as those idyllic selves (a big problem for people looking to undergo positive transformative change). Lastly, we can work progressively toward those goals.

Next Steps

Further study should focus on best practices in psychology and phenomenology. Psychology for realizing reasonable boundaries and implementing systems that are, perhaps, incapable of producing results past those realistic boundaries. For instance, a person who is 4’5 as an adult is not going to wake up one day 6 ft. tall. Phenomenology because it is important to keep the “lived body” in mind and recognize the unique experiences and challenges of various kinds of lived bodies, making sure not to stigmatize what a person is today but still encouraging them to work toward the realistic representation of what they want to be. It’s a tricky balance, idealistic perhaps, but I think there is potential.

Also, there are two major shortfalls of "The Proteus Effect" study and subsequent studies seeking to expand or build upon that research. First, all studies fall to show any sustainment of the attitudes adopted by people from avatar usage. The interactions recorded were always directly after immersion in a virtual environment and usually lasted only 20-30 minutes. There may be an assumption that people prone to such behavioral changes are so frequently involved in virtual environments that the effects could be sustained through repeated use, still that assumption warrants challenging.

Second, the Knutzen and Kennedy study did not disaggregate the data related to how people rated avatars. Much in the same way De Saussure (2011) describes a signifier and signified, the latter being assigned by a myriad of recollections and learnings, physical traits can work that way as well. What if blonde hair was a signifier of "attractive woman" to a participant where brown hair is not? Those sorts of distinctions are equally as important to understanding how we perceive and manipulate our digital selves.

As it stands right now, I think the best approach with regards to my daughters is the approach I’ve been taking. In as neutral a voice as I can muster, I ask them questions about virtual representations of themselves. I ask my daughter why she altered her eyes the way she did? Why did she apply the make-up she did. Why is her iDressup avatar dressing in the style that she does? How do they think that style of dress uniquely equips them to present themselves to the world?

We have conversations like this quite often and I find that the avatar, as a mechanism of the virtual world, that impacts our real world, is a powerful tool in better understanding the desires, needs and frustrations of my daughters.

Bibliography

Avatars for Behavior Change. (2013) Accessed online December 4, 2013. http://pie.eng.usf.edu/avatars4change/

De Saussure, F. (2011). Course in general linguistics. Columbia University Press.

Diedrich, L. (2001). Breaking down: a phenomenology of disability. Literature and Medicine, 20(2), 209-230.

Fox, J., Bailenson, J. N., & Tricase, L. (2013). The embodiment of sexualized virtual selves: The Proteus effect and experiences of self-objectification via avatars. Computers in Human Behavior, 29(3), 930-938.

Knutzen, K. B., & Kennedy, D. M. (2012). Designing the self: the transformation of the relational self-concept through social encounters in a virtual immersive environment. Interactive Learning Environments, 20(3), 271-292.

Lovink, G., & Nakamura, L. (2005). Talking race and cyberspace: An interview with Lisa Nakamura. Frontiers: A Journal of Women Studies, 26(1), 60-65.

Thomas, A. G., & Johansen, M. K. (2012). Inside out: Avatars as an indirect measure of ideal body self-presentation in females. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 6(3), article 3. doi: 10.5817/CP2012-3-3

Van Der Heide, B., Schumaker, E. M., Peterson, A. M., & Jones, E. B. (2012). The Proteus Effect in Dyadic Communication: Examining the Effect of Avatar Appearance in Computer-Mediated Dyadic Interaction. Communication Research.

Yee, Nick; Jeremy Bailenson (2007). "The Proteus Effect: The Effect of Transformed Self-Representation on Behavior". Human Communication Research 33: 271–90.