An Equation for Adaptation of Organic Ecological Learning Machines

In life, there is good pressure and bad pressure. When a new food resource becomes available, that's good pressure that drives a population upward. Conversely, scarcity and disease apply downward pressure on populations. Within any population, there will be variations that lead to the rise of one sub-group, and the fall of another.

Let's use an example that lends itself to formulating an equation.

We're studying change, which happens as a result of pressure, or P. No pressure means that P = 1 and any value multiplied by 1 does not change. If there is no upward or downward pressure on a population then its level of energy utilization stays the same. Completely static situations in nature are rare, but as long as the P-value hovers around 1, there will be relative stability and no obvious changes or disruptions. If Y represents our population measured by its respiration/calorie/energy use, then if P = 1, PY = Y.

Now let's say that Y represents energy utilization or "flow" of a population of Proteobacteria living in your gut, a harmless species that nevertheless suffers negative pressure represented by a P value of less than one when you take an antibiotic to counter an infection unrelated to it. Collateral damage, right? Some of Y has resistance to the antibiotic, and their population and flow increases, becoming Y1. The rest of Y are unable to metabolically break down the antibiotic, and so they become Y2 which goes into decline as the population succumbs. Once Y2 and the other species are gone, there will be a huge opportunity for Y1 to multiply, and so its P-value will go through the roof. To keep it easy, we'll call Y1s new P-value P1. Meanwhile, Y2s P-value is dropping to near-zero levels as the population goes extinct, so we'll call Y2s new P-value P2.

Therefore our basic equation for a population's binary response/adaptation to change as a result of a variable representing the sum of environmental pressures is:

PY = P1Y1 + P2Y2

For the above to be true P cannot be 1, and Y1 and Y2 cannot be the same, or there would be no divergence in the population. The population that is more successful will be represented by Y1, and the population with the less successful flow will be Y2. Although we would expect that the P-value for Y1 (P1) would continue to be higher than the P-value acting on Y2 (P2), that is not always the case. Going back to our Proteobacteria example, there is a cost to lugging around the machinery to break down the antibiotic that is not worth it when there is no medicine to resist. If the application of antibiotics continues, as it would in animals regularly consuming feed laced with antibiotics, then P1 will triumph and P2 will disappear. If antibiotics are only applied once, then P2 will dominate and P1 will disappear. The equation will run, and run again, giving a snapshot of who's using how much of the total in the niche. New branches of population flow will emerge, with many branches disappearing.

As mentioned at the outset, pressure can be good or bad. Population flow can vary in terms of positive response - some groups can really pack it in at the all-you-can-eat buffet. On the other hand, constant downward pressures due to pollution or habitat loss lead to negative P-values for all and a diminished overall environmental carrying capacity in which all population flow diminishes. Running the equation over and over could be predictive. A visual model of divergent populations as a "Y" on its side is an intuitive way to show an initial flat baseline, then the application of pressure leading to the rise of one successful sub-population and the decline of another sub-group. This reflects the reality of natural systems.

While some might say that this is natural selection in evolution, I would disagree. An organic learning machine does not produce random mutations that are occasionally successful. An organic learning machine comes pre-programmed with responses to change that are stored in various sectors of the population's genome, and can be shared and used to advantage in response to environmental stress. My equation is therefore quite different from the Darwinian model, and far more testable within a framework of improved understanding of genetics and models of ecological population flow.

It could also lead to increasing complexity, as part of what I like to call "the starter pack".