Useful Doctrines from Cybernetics and Cognitive Science

Below are the philosophies, principles, and hypotheses I find essential when researching human-level AI and translating cybernetics and cognitive science into machine learning.

Philosophy

1. “What I cannot create, I do not understand” (Feynman)
   To truly understand a phenomenon, one must be able to reconstruct its mechanism from the ground up.

2. “What is it like to be a bat?” (Nagel, Uexküll)
   Non-human qualitative experiences—such as a bat’s echolocation—can never be fully captured by a human third-person objective description. Jakob von Uexküll’s concept of the Umwelt highlights that each species’ unique perceptual world never completely overlaps with another’s.

Hypotheses / Principles

1. Bayesian Brain Hypothesis (Helmholtz, Hinton, Doya)
   The brain treats sensory inputs as probabilistic models and updates them according to Bayesian rules, building on Helmholtz’s notion of unconscious inference.

2. Free Energy Principle (Friston, Parr, Pezzulo)
   Living systems minimize variational free energy to reduce prediction error and maintain stability. Parr and Pezzulo extended this into the theory of active inference.

3. Emergence, Holism, and Gestalt Principles (Koffka, Bertalanffy, Anderson)

   • “The whole is more than the sum of its parts.” (Kurt Koffka)
   • Systems holism (Ludwig von Bertalanffy)
   • “More Is Different.” (Philip W. Anderson)
     Complex systems exhibit novel properties that cannot be deduced solely from their individual components.

4. Computational Irreducibility (Turing, Wolfram, Chaitin)
   Certain processes cannot be predicted without simulating every step of their computation, as shown by the halting problem, algorithmic randomness, and cellular automata research.

5. Law of Requisite Variety (Ashby, Boisot & McKelvey)
   To regulate external complexity, a system’s internal variety must match or exceed that of its environment. Organizations must design both the quantity and structure of their internal complexity to meet environmental demands.