Complexity Theory
Introduction
Behavioral environments are complex by nature and worthy of the application of Complexity Theory!
-----------------------------------------------------------------------------------
-----------------------------------------------
The word 'complex' is visible throughout this Web site. The reason? We're dealing with skills and their normal operational environments (virtual realities), which are very complex, and we are engaged in building realistic representations of those complexities. The complexity may seem more obvious when our involvement is with organizational skills, since the interactions among the organization members have to be considered, but the environments of personal skills can be just as complex and involve many interactions. See tennis, for example. See also Context.
Our main concern is with the dynamics in those environments. So it's natural to take seriously any theory bearing on the general nature of complex systems. Hence we look at complexity theory, or the theory of nonlinear systems, as it's otherwise understood.
Here we look at the work on Chaos Theory I started in Simulations. This takes us into matters of order, stability, computability, and predictability. In Simulations I cited the system widely known as The Beer Game, which can quickly go unstable. Here, we consider environments such as the stock market and sports.
-----------------------------------------------
Complexity theory is a field of study that has grown out of investigations into turbulence and characteristics of structures exhibiting fractional dimensions. The investigations led to more mature theories, Chaos and Fractals, which have since been generalized to Complexity theory.
In his book, Chaos and Order in the Capital Markets, Edgar E Peters discusses the nature of complexity as follows:
Over the last few years, it has become clear that chaos theory and fractals are a subset of a much larger universe of discourse: complexity theory. Complexity theory deals with processes where a large number of seemingly independent agents act coherently. Complexity can be a dynamical process, or an object. We recognize the qualitative aspects of complex objects without being able to precisely measure them. Trees, handwriting, and riverbeds are all objects that are individual yet have global characteristics. Complexity assures that they are different in detail while similar in concept. That is, they are locally random, but globally deterministic. They are
fractal.
---------------------------------------------
Our skills arenas, or virtual realities, more than meet the minimum requirements for admission to the class of complex objects, both as to their structure and their dynamics. For one thing, their components are interactive -- they affect each other in diverse ways, short and long term, at times even well into the future. There are frequently many feedback loops in the environments. And the arenas exhibit a significant degree of adaptability. This often leads to chaotic states.
Also, the more closely the environments are examined, the more detail there is to be found, seemingly without limits. There are literally multiple layers of complexity. Just look at the structure of the human organism, with its many bones, muscles, and nerves -- not to mention the billions of neuronal cells in the brain. Prices in the stock market also have multiple levels of complexity. This is characteristic of fractals, which have become an integral part of complexity theory.
Given the interactive nature of components, you can see that the dynamics are non-linear and yet more complex -- particularly if you take into account the perceptual aspects of the subject in the environment. When it's recognized that the individual is only part of the larger arena, the complexity of the latter can be seen in its true proportions.
---------------------------------------------
The word nonlinear literally means not linear. In behavioral environments this is to say that actions generally don't lead to proportionate or straight-line reactions. In a linear framework, if I were to modify my efforts by a certain amount, I would change the results proportionately. That is, there is no recognition of the current state -- no memory, which could influence the current response. Click here for details.
-----------------------------------------------