 
|
In this simulation an organism consists of a genome containing the propensities to turn in various directions. The organisms live in a toroidal universe and feed on a continuously raining supply of food. If you've searched for artificial life information, you've probably seen this type of system before.
When I described this system to a friend of mine, he was intrigued. He later came back to me with the two images at the top of the page. The image at left is some basic bacterial culture. The image at right is some basic bacterial culture in which the bacteria have reached the nuclear age.
I've written several variations with various methods of data display. There are a few things that can be learned about this system.
This image shows the population data from twenty identical simulations. The x-axis represents time
and goes from 0 to 50,000. The y-axis represents population and goes from 0 to 100. The data is averaged
across 1,000 time steps to produce this image.
From this you can see that while the early history of the system can be readily predicted, it rapidly develops into a chaotic state where you are unable to make detailed predictions of how the system will develop. An equilibrium state is reached which includes chaotic oscillations around a median population level. |
This image shows the population data from thirty simulations, ten for each of three conditions. The x-axis
represents time and goes from 0 to 50,000. The y-axis represents population and goes from 0 to 100. The data
is averaged across 1,000 time steps to produce this image.
Each set of data is represented with a different color. The simulations differ in the rate of food addition to the environment. The food addition rate in the red simulation was half that in the green simulation. The food addition rate in the blue simulation was twice that in the green simulation. The equilibrium population level in the red simulation was half that of the green simulation. The equilibrium population level in the blue simulation was twice that of the green simulation. (The same results are produced when the value gained by an organism from eating food is doubled or halved.) It apparent from this set of simulations that more food (or more food value) allows the population to develop to higher levels and to reach an equilibrium state more rapidly. In this system, over the range of values tested, it was also apparent that there is a direct relationship between food input and resulting population level. |
This image shows the population data from ten simulations, five for each of two conditions. The x-axis
represents time and goes from 0 to 1,000,000. The y-axis represents population and goes from 0 to 50. The
data is averaged across 10,000 time steps to produce this image.
In the five sets of simulation data depicted with green, the organisms mutate. In the five sets of simulation data depicted with blue, the organisms do not mutate. The mutating populations reaches a level consistently higher by about five organisms. The mutating populations are able to reach a higher density because they are evolving. The individual organisms become more efficient in their search for food over time. |
This image shows the population data from ten simulations, five for each of two conditions. The x-axis
represents time and goes from 0 to 50,000. The y-axis represents population and goes from 0 to 100. The
data is averaged across 1,000 time steps to produce this image.
In the five sets of simulation data depicted with blue, each new organism is initiated with the same initial energy level. In the five sets of simulation data depicted with green, each new organism is initiated with a random initial energy level. The average initial energy level is equal for both sets of simulations. No difference in results is apparent. |