I bought my girlfriend a Wii some time ago and before playing any game we spent 2 hours making Mii's of ourselves and all the people we know. A Mii - for those not in the a know - is your Wii avatar. It is associated with your personal stats on games such as WiiFit and WiiSport. Now Mii's are downright cartoony, but we tried to make them as lifelike as possible. After you're done it asks for your weight and height (for WiiFit) and calculates your BMI. Now I had made my Mii a fairly fit looking character, but given I carry a bit of muscle, my BMI came out as 'overweight' and it updated my Mii accordingly, and blew the little guy out. I felt outraged! That is not me, what I created was me! And it wouldn't change me back without a judgmental message (yeah, WiiFit totally judges you).

I also created a second Mii. It was my stoner alter-ego. I used him exclusively when I was playing on the Wii after drinking. I created a dummy account because I didn't want to skew my 'real' stats.

I've also created some kind of weird cat-class avatar (who might have also been female, I can't remember) when I played Morrowind: Elder Scrolls (a game which I probably invested 150+ hours in).

What does it mean? Why did I protect the real Mii against drunken Wii-Golf stats? Why do all nerds on youtube with character-class tutorials have hot female avatars? Is the warrior really better than the caster?*

Trepte and Reinecke (2010) suggest that a number of factors come into a play when players create an avatar. Those factors include competitiveness of the game, Identification with the Avatar, and Self-Satisfaction. Intuitively we could probably list off a few more - gender and sexual orientation spring to mind, and I'm sure there are more (beyond the confounding class-benefits of stats**). Trepte and Reinecke (2010) found that the less competitive a game was the more similar our avatar was to ourselves in terms of personality characteristics, and the more competitive a game was the less similar an avatar was. Additionally, those who were less satisfied with their own personality characteristics created a character that was more dissimilar to themselves than those who were satisfied. 'Identification' encompasses the 'connection' between player and avatar, and the intentions of the players regarding the goals of the game. On occasion I have literally spent an hour crafting my perfect character before setting out, only to go back, recraft, and begin again because my avatar didn't 'fit' me. I'm sure I'm not alone in this experience. They found that identification boosted enjoyment with the game, particularly for non-competitive games.

...yet it's worth skipping the Freudian crap. Trepte and Reinecke (2010) report that competent and satisfied people create competent and powerful avatars, and less competent and less satisfied people create competent and powerful avatars. The trick, perhaps, (which they acknowledge) is not so much what people are thinking and how they are representing themselves, but how they intend to play. It is exceedingly reasonable to suggest that players are simply optimizing their playing experience - who wants to handicap their playing experience? +3 to attack - Yes, Please. -2 to defence - No, Thank you. That's not to say a carefully controlled study can't reveal some really cool findings, but the quasi-experiment that Trepte and Reinecke (2010) conducted doesn't quite cut it.

I used to play Diablo II on Battle.net. You'd rock up to a game, trade a bit, head out and kill some demons, then rinse and repeat. For hours. It was great fun. But you also had the option of cutting someone down, stealing their gold and claiming their ear as a trophy (a risky strategy), retreating in battle, letting them fall, and claiming their gold and items (less risky, but a serious gaming faux pas), or helping unconditionally for mutual benefit (common and fun). All worked well, but the second one was definitely the best for quickly earning cash. Smith (2010) has a theoretical paper regarding trust in games and explains some existing models, games and examples. He contends that there are genuine cues as to a players trustworthiness in game - be they imposed by the game design, or organically emerged - that allow the game to be successful. He doesn't appear to address in-group / out-group dynamics, which might be a genuine mechanism in deciding trust. For instance, I might choose not to enter the battle field if there's a player that could overpower me in seconds, but rather choose someone like myself (similarly ranked) such that we both benefit from working together evenly, and any battle between us would be difficult and costly to the both of us.

So a lot of games come down to zero-sum equations and prisoner's dilemma type scenarios. The above is a classic example, but with the exception of the trophy ear, it wasn't the focus of the game. Here's one that builds in a whole new level Game Theory. The premise is that you and a crew of online buddies conduct a heist. At this point if you're successful in escaping you all split the cash evenly. But at any time you can top your buddy, add his share to the pool and keep going. I'm pretty sure there's some bonuses for co-operation, but the gains are quicker and easier if you play the traitor. The game, naturally, becomes a uniquely tense experience where you're not so sure if it's worth shooting your mate in the back before he does so to you.

In my mind, however, there's nothing ground-break nor experimental within [Smith, 2010] , so I won't go into any depth. I remember reading anecdotal reports of this game at the time, and it seems that Game Theory pays off, with the great majority co-operating all/most of the time, with only a small percentage of traitors.

...and of course, the only question left to ask, is it possible to behave morally in a game that is built around such a well research principle (game theory?).

*Obviously not. That's ridiculous.

* My cat class character benefited from some 1337 stealth skills....

---

Trepte, S., & Reinecke, L. (2010). Avatar Creation and Video Game Enjoyment Journal of Media Psychology: Theories, Methods, and Applications, 22 (4), 171-184 DOI: 10.1027/1864-1105/a000022

Smith, J. (2010). Trusting the Avatar Games and Culture, 5 (3), 298-313 DOI: 10.1177/1555412009359764