SOMA (self organising map ants), produced 2005
SOMA is a software model of a dynamic system in which a virtual population of ant-like drawing agents develop individual behaviours in response to marks on a surface which they collectively modify.
The motion of each ant is regulated by a small neural network called a Kohonen self organising map (SOM) which responds to nearby patterns on the drawing surface. The SOM determines the ant's next move and is modified by the most recent pattern in the process. The ant leaves a trail, contributing to the overall image.
With thousands of ants interacting this way, it's a complex multi-directional feedback system in which agents indirectly influence one another's internal structures through the effects they have on their surroundings.
It's difficult to predict what will happen without running the system. These images are taken from separate runs, with different starting configurations, numbers of ants, trail persistence, sensor configurations, movement capabilities, individual trail colours and colour responses. I had a bias in seeking to find combinations of parameters which would lead to clear yet constantly changing visual structures while avoiding sequences which either dissolved to grey or seemed to get stuck in a rut.