Experimental mechanic (context-free)

[joe_llama]

Here's a half-baked idea that fell on me last night just before I went to sleep. It's quite possible someone else thought about it already, so I apologize in advance.

I want a mechanic that allows a character to be dynamic, to change with time (not just advance), to have a flexible behavior much like real people have. People change their opinions and goals more than a few times throughout their lives, especially people in stressful or condensed situations. I think game characters should have this quality. I also want characters to act out of their desires, ambitions and feelings instead of their skills and abilities.
 
IMO, each character is made of five essential parts: Background, Personality, Motivations, Viewpoint and Relationships. The first three represent the character's past, present and future. Viewpoint and Replationships can actually be "derived" parts from these basic three.

Partially inspired by neural (learning) networks, the Pool, and Narrtivism in general, I came up with this mechanic:

Each character has two lists: Personality traits and Motivations. They are drawn from the character's story (his Background).

Each Trait/Motivation is rated by a dice pool. Whenever a character tries to do something, it chooses one Motivation plus one Trait appropriate to the situation and rolls that much dice to determine success/failure in a Pool-like fashion. It's possible to use either a Trait or Motivation if there is no relevant "partner" to go along with it (smaller pool though). If there is no relevant quality, the GM determines the outcome of the event (For now. Possible change in the future).  

For every success, the Trait/Motivation is increased by one. For every failure, the Trait/Motivation is decreased by one.

When a Motivation reaches 0, the character no longer feels the need to follow that path and may choose a new motivation to replace it. When a Trait reaches 0, it is replaced by an opposite, semi-opposite or other appropriate personality Trait.

This is what I have so far. I have a lot of questions and thoughts but I'd rather not say them out loud yet. What do you think? Can you see where this mechanic breaks down or in need of fixing? Any kind of opinion is welcome.

With Respect,

Joe Llama

[Mike Holmes]

Well have you looked at Synthesis? I designed it to do a lot of what you're thinking. There isn't a two layer split, and it's not all internal characteristics, but it can be used that way. See if it helps at all with crystalizing your idea.

Mike