Monday, October 08, 2007

Neuroticism is optimal?

I just read this article:

Neurotic software has winning personality

Austrian researchers made several different AI classes (agressive, defensive, normal and neurotic) and pitted them against a real-time strategy game's built-in AI engine. Turns out the neurotic AI was the most successful of them all

In other news,

Lately I've been watching a lot of Johnathan Blow's speeches, where he often mentions "unethical gameplay" (article). He defines unethical gameplay, in one specific case, as designing a game with "rewards" built in. So, the user keeps playing a game for these moments of reward (cutscenes could be one example) while the underlying gameplay is crappy or uninteresting. He says that the play that results from the game should be interesting in of itself, not because of the "rewards" that the game doles out.

Both of these items are interesting to me. While the majority jury is out on whether violent games cause violence (I don't think so) I am of the belief that the cognitive structure of a game has a significant effect, in the short term at least, on how you behave. After playing a few hours of the Thief series, I find I always stick to the shadows in real-life for a while afterwards. After playing a "twitch"-style casual game for 15 minutes, I have a little trouble concentrating and feel hyper afterwards. But these are always short-term effects. The violence in videogames is different because it is representational. I'm not going to get deep into this right now, but you aren't actually ACTING violent, you are DOING violent things. If you're doing them with a friend, such as completing Halo 3 on Legendary (me and my housemate yesterday) then the primary behavior that is being encouraged is cooperation.

So, if game designers want to design ethically, they should think of the type of personality that a game encourages. Cool.

No comments: