Efficient Competition

Efficient competition among intelligent opponents requires irreducibly uncertain behaviors. John Maynard Smith said that biologically we should see this uncertainty within and between individuals. However, it seems interesting that game theory provides us only with a theory of static equilibrium in games among two or more players. What about decisions we make (games we play) that seem to involve only ourselves? Do they always involve other parties, do we model "the world" or "nature" as the other player? These questions seem unanswered under my current understanding, this could be wrong as I gain a better understanding of game theoretic research, but nonetheless they are interesting ideas.

The implementation of a mixed strategy in decisions made by ourselves seems a difficult concept to grasp. We must find a neuronal basis for randomness, a basis that shows the fundamentally stochastic nature of our brains that allows us to implement algorithms that require mixed strategies and other probabilistic behaviors. I have written on this before, but I believe more work needs to be done to demonstrate this principle unequivocally. Our intuition would make one believe that we are not making decisions randomly even when we might actually be doing just that.

As intelligent agents we must develop algorithms to exploit structure in the environment. If any pattern exists that underlies the world around us, then those who can exploit this most efficiently will be more likely to promote their inclusive fitness. Henceforth, evolution is driving our brains and behaviors towards optimal pattern recognition and causal linking abilities. Neuronal predictions of the future and decision making abilities are directly tied to our ability to see cause for effect and store this information in our cognitive map of the probabilistic nature of the world. The only problem with this is that I do not believe people are that good at understanding causal relationships. From a general and unscientific point of view, it just seems that people tend to see what they want or expect and not necessarily the true relationships. This must be tested experimentally and if confirmed, must be explained analytically.