The introduction of noise into a model provides another example of the interest in between. In noiseless systems, agents quickly get stuck, often in inferior configurations. With a lot of noise in the system, chaos reigns, and little, other than frenetic movement, is possible. Adding just a bit of noise to a system, however, often induces order and leads it toward optimization.
