An Alternative Visual Model for a CAS

Complex Adaptive Systems are complicated to model and to understand. 

The application of a model like this to a particular entity like a social system or an ecosystem or an economy can be confusing, but typically you'll find common ways to do it with in any particular arena.

The nice thing about this model is it's basically a dissection of the overarching process. It's just that we don't normally need to see the insides of all this to be able to use it.

One of the downsides of the above is that it underplays the impact of environment, which serves a primary, not secondary role in shaping CAS behavior.

Furthermore, the above model doesn't say anything about the roles of the individual agents, even though they can be broadly categorized into at least two discrete groups.

Another issue with the model above is it's oriented as though the "complex adaptive behavior" is what we're trying to figure out. In most cases, we already know the emergent behavior, and we have some idea of the local relationships. What we're usually looking for then, is either the moving parts that connect one to the other, or often as not, simply plotting a course based on past behavior.

The former is difficult no matter what. The latter is simple, as CAS follow a simple overall dynamic where this is concerned. CAS have inertia due to the positive and negative feedback loops

I propose a very simple alternative model when you don't need the inner workings of the CAS, and you know the emergent behavior with some idea of the local relationships. This alternative can be useful in terms of charting the overarching course of a CAS, and locating it within the downcycle or upcycle phases of its evolution.

Here's the most basic example

In the context of modeling economies, agents would be market transactions (or if you want to try for it - market actors, though that can get dicey) - the predictable ones versus the disruptive ones. When a market tanks, the sell offs of previously profitable stocks is disruptive. As the market tanks there are greater numbers of these sell offs and they encourage more sell offs, taking into account that negative feedback inertia, up until it reaches an equilibrium again. That process is the downcycle phase. The high water mark rises, tapers, then crashes usually slowly rising in response to a market tanking and finally recovering.

In the context of modeling societies, agents would be individual social behaviors (or if you want to try it, social actors) - the predictable ones versus the disruptive ones. A crime is disruptive. Response to a crime is disruptive. As crime increases, and response to it increases, the high water mark raises until it finally finds its equilibrium, often again with a finally drop as said crime is now once again "under control" - until conditions create the rise again and crime reasserts itself.

This is very simplistic, but I'd argue it's very useful as CAS go because it makes understanding the motion of them much easier.

I've never really tried to flesh this out before, I just keep this stuff filed away in my head. Any attempt to articulate it is a work in progress.