Six Structural Attractors Unify Computation, Artificial Intelligence, and Biological Consciousness

By compressing 203 algorithms from 20 computational domains to their Kolmogorov minimum, six structural attractors emerge: PARTITION, ACCUMULATE, RECURSE, SEARCH, TRANSFORM, ENUMERATE. These are argued to be irreducible mathematical primitives of computation. The authors propose the PAR Consciousness Hypothesis, contending that the triplet {PARTITION, ACCUMULATE, RECURSE} is the minimal necessary and jointly sufficient substrate for consciousness, mapping to Integrated Information Theory, Global Workspace Theory, and Higher-Order Thought theory. The work offers 24 falsifiable empirical predictions and 5 architectural predictions for AI systems, claimed to be testable within 6–12 weeks, positioning it as a computationally grounded, falsifiable theory of consciousness.

Study at a glance

Abstract

We report the empirical discovery of six structural attractors — PARTITION, ACCUMULATE, RECURSE, SEARCH, TRANSFORM, ENUMERATE — that emerge when 203 algorithms drawn from 20 computational domains are independently compressed to their Kolmogorov Complexity minimum via multi-strategy search. These attractors constitute the irreducible mathematical primitives of computation and provide a unified framework connecting algorithmic information theory, artificial intelligence architecture, and biological consciousness theories. The flagship result is the PAR Consciousness Hypothesis: the triplet {PARTITION, ACCUMULATE, RECURSE} constitutes the minimal necessary and jointly sufficient substrate for consciousness, mapping directly to the three leading neuroscientific theories — Integrated Information Theory (IIT), Global Workspace Theory (GWT), and Higher-Order Thought theory (HOT). We provide 24 falsifiable empirical predictions and 5 architectural predictions for AI systems, testable within 6–12 weeks, making this the first computationally grounded, falsifiable theory of consciousness.