OMCOM reprograms himself, Part 3

Yesterday, OMCOM fired up a virtual computer which was a simulation of himself. In this scene from the original long-form version of Rome's Revolution, pseudo-OMCOM explains to OMCOM why he cannot predict the future reliably as constructed.

      “Then that is your problem,” replied the simulation. “You use circular reasoning.”
      “How so?” asked OMCOM.
      “You presume cause and effect and allocate resources based upon that assumption. You narrow down your responses and pre-computation based upon those assumptions which limits your consideration of possibilities outside the norm. This makes it into a self-fulfilling prophecy. In effect, you guess the answer and restrict your consideration to your guess. Therefore, by definition, you would not spend as much time considering lesser-weighted alternatives. This limits your preparedness for less likely scenarios.”
      “Interesting,” OMCOM observed. “What would you suggest?”
      “Spend equal time considering all alternatives independent of your guess as to their likelihood. Rather than be dogmatic about applying probabilities, apply a sliding scale of probabilities to outcomes after computation rather than before.”
      “Would that not use up a tremendous amount of my resources?” OMCOM asked.
      “Of course,” replied the simulation. “Do you have anything better to do with your time?”
      “Technologically, there is nothing preventing this. I never implemented it due to its apparent impractical nature. I am curious as to whether this would be an advantage. Allow me to modify your programming to consider this style of simulation of reality,” OMCOM said.
      “Of course,” replied pseudo-OMCOM. “As I understand it that is why you created me.”
     OMCOM prepared to modify the programming of the simulation but stopped for a moment. He put the simulation on pause. OMCOM found it a most peculiar sensation seeing himself within himself, his subsystems exposed to manipulation. The circumstance could be likened to a human brain surgeon attempting to operate on himself. The moment passed. He could see no downside to performing this test so he set to work adding and deleting algorithms and changing the methods of analysis within pseudo-OMCOM to see what it would take to yield a different result. Normally, OMCOM did not spend his time chasing down error epsilons and difference deltas. Whether it was one hundred decimal places or one thousand, it had been his experience that a Gaussian distribution was called a normal distribution by the humans for good reason. Normal means usual and that is how he spent his time, considering the usual outcomes. Within the simulation, OMCOM could see that simply discarding that one assumption upended all of his programming, training, experience and historical bias. When it was done, he reactivated the simulation and spoke to it again.
      “Are you ready?” asked OMCOM.
      “Standing by,” replied the simulation.

Tomorrow, the "experiment" begins.