System complexity is difficult to quantify let alone study, which gives rise to the adjective used to describe it: “complex”. The difficulty may stem from the fact that the system is enclosed, such as a vertebrate’s brain and/or there are too many changing components of which to effectively keep track and measure.
Chris Adami of Michigan State and his colleagues designed and created a complex system that has the property of being easy to access for study purposes, to observe the behaviour and evolution of such systems. They did it by writing a program which generates a swarm of artificial, virtual organisms which they called animats.
Here is a diagram of one:
The task given to these creatures was to navigate their way through a virtual ‘maze’ to find doorways and pass through sections, where at the end, the ones that made it could reproduce. This seemingly elaborate genetic algorithm produced an emergence of ”memory”, formed over generations of reproducing animats.
Source code for Markov brains: https://github.com/dknoester/ealib