We have mentioned above the link between the idea of a controlled switch and the idea in evolutionary biology of a selection criterion and process. The interesting thing our experiments bring out is that the dynamics of iterating systems containing controlled switches can be very irregular and unpredictable. If we regard the controlled switches as altering the system structure, it is clear there is application to biological systems since the ecological environment is set partly by the densities of the populations which inhabit it, and so there is a direct feedback mechanism. The analogy is less clear cut however, since by its nature a switch is an on-off or logical element, and in ecological terms the structure variations are more gradual. Therefore a more accurate and useful model might be made by having a more-or-less linear feedback mechanism to describe ecological evolution, with the addition of controlled switches to model catastrophic population-level induced effects. Such a model might be better at explaining the relative robustness and stability of ecosystems, while at the same time providing for the occasional unexplained sudden changes which can occur.
Viewing the model of evolution suggested by these electronic experiments as having the selection process of a controlled switch driven by variability which is really a chaotic process in the dynamics of the system, we see that this model suggests that the variability does not have to be truly random, in the sense that it occupies the entire system state space with uniform probability; sufficient variability is present in the chaotic attractor (occupying a small comparative volume of the state space, even possibly of lower dimension) to drive the selection mechanism. Thus if the chaotic variability coupled with selection results in favourable emergent behaviour, we can envisage a process leading to more rapid evolution than would be the case if the variability was by chance.