In the simulation below, there are three species of organisms: stars, plusses, and pentagons. Their behavior is controlled by parameters that you can control with the
sliders below. For each species, there are three parameters that determine how quickly it moves away from (negative values) or towards (positive values) each of the
species, including itself. In addition, each organism has a little bit of randomness added to its movement. There are three other parameters that control the overall
amount of movement of the organisms, how much organisms tend to move in the same direction they have been moving (inertia), and how rapidly the influence of other
organisms falls off as the organisms get further away. You can also pick up, move, and "throw" individual organisms by clicking on or touching them. There are some
preset configurations that you can explore. If you come up with an interesting pattern, please share the parameter values with me.
Movement of Plusses:
Movement of Stars:
Movement of Pentagons:
Falloff of Influence with Distance:
This simulated ecosystem shows a number of very common phenomena associated with systems. As you explore it, some general system phenomena to look for are:
The value of randomness: Does the randomness added to the organisms' movements seem to help them get unstuck in some positions? For example, in the "segregation"
configuration, once the species have separated into different groups, take a plus from its group and move it to the middle of a group of pentagons. Notice how long it
takes for it to return to its species' group. Now, do the same thing again, but set the "inertia" parameter to 0, which will make each organism's movements more random from
moment to moment. Did setting inertia to 0 speed up the time required for the plus to return to its group? Injecting randomness into a system can help shake it out of stuck patterns.
Do the same parameters sometimes lead to very different patterns depending on random differences in the starting conditions? For example in the "rock scissors paper"
configuration, you may notice that sometimes the organisms circle around clockwise, sometimes counter-clockwise. In physics, this is known as a bifurcation.
In a positive feedback cycle, increasing something leads to even more of that something. For example, views of a short movie on a video site may lead to
even more people watching the movie if users notice that the movie has been viewed by many other people, become curious about it, and decide to watch it for themselves, thereby
increasing its view count for the users who visit the site later. The "meiosis" configuration has a good example of
a positive feedback cycle in which as soon as there is a small segregation of stars from crosses it leads to more and more separation. A characteristic feature of positive
feedback cycles is that a process that starts off slow gets increasingly rapid.
Self-organization: systems often create highly structured patterns even if there is no parent, teacher, president, or boss giving orders. Sometimes the simple
repulsions and attractions between the organisms lead to large, distinct patterns for the entire group. The "segregation" configuration provides a good example of this
in which the tendency of an organism to move toward members of its own species and away from members of other species leads to large and separated groups in which
the members of each group belong to the same species.
Emergence: Related to self-organization, "emgergence" refers to situations in which local interactions between elements give rise to fundamentally different global properties.
In the ecosystem simulation, the parameters that determine how much each organism attracts or repels every other organism lead
to very different properties at the level of the entire population. For example, in the "egg" configuration, almost every time you reset and run the simulation,
the organisms will almost always settle into a core of yellow pentagons, an outer shell of stars, and then two groups of plusses at the ends. Why do the rules of
attraction always lead to the emergence of exactly two groups of plusses?
Adaptation: Systems are often highly adaptable, robustly "fixing" themselves when they are pushed out of their natural state. For example, in the "triplets"
configuration, after a minute or two, the plusses will be separated from each other, effectively covering the entire space. So will the stars and pentagons.
In fact, you may notice many plus-star-pentagons triplets. If you move several plusses to the same location, they will quickly adapt so that they become well
spaced out again.