Imagine a large, floating, black torus. The torus is initially covered at random by thousands of small red dots, so that little of the black surface can be seen. These dots are simple creatures that follow only one rule. Each of them chooses at random one other dot, and then moves in the opposite direction of that one dot. They move in time steps; the distance they move each step is inversely proportional to their current distance from the one they are focused on eluding.

Each of these thousands of dots move in parellel. Each only knows of one other dot. There is no overseeing coordinator of the movement of the population of dots. Many would suspect that the collective movement would be simple and chaotic. One dot avoding another, which in turn moves because of the location of another, and so on. Each dot could run around the torus it sits on forever trying to reach the location farthest from the one other dot they know about.

But, as unlikely as it may seem, complex order does arise from the simple, decentralized actions of each dot. They agregate into a small number of "Quantum Life Forms" (QLFs).

The simulation of this world was programmed using StarLogo, which can be downloaded free from the StarLogo site. In the words of StarLogo creators, "StarLogo is a programmable modeling environment for exploring the workings of decentralized systems -- systems that are organized without an organizer, coordinated without a coordinator."

Below are a series of time-lapse images of the simulation in action. You can click on any of these smaller images to see them in their full size.

The sides of the space wrap around to the opposite side, to form the torus mentioned earlier. These pictures show the initially random placement of dots forming into a small number of coherent figures, the QLFs.

They have been named Quantum Life Forms because each one has an identical twin. But unlike natual twin, who are identical at birth but who differ through time in their actions, these QLF twins are always exactly the same. Each QLF will flow, swirl and move across the landscape, always ramaining the same as their twin. Even if the simulation is paused and some of the elementary dot elements of a QLF are moved away, the QLF will reform quickly to match its twin after the simulation is restarted. Also, the twin of the QLF that was disturbed will also show disturbance, even though it wasn't modified!

A real sense of this phenomenon can best be gained from watching the simulation, and even setting some critera yourself if you download the simulation. The simulation must be run using MacStarLogo, not the newer java based version.

This simulation is only one small example of how simple, decentralized actions can lead to organized, emergent behaviors on higher levels. The StarLogo site has many more examples, as well as many more artificial life explorations on the web.