Intelligence as high-dimensional coherence: The observable dimensionality bound and computational tractability.
Bio Systems February 1, 2026 Ian Todd
Intelligence depends on maintaining high-dimensional coherent dynamics. High-dimensional systems are effectively nonergodic on relevant timescales, confining trajectories to tiny regions of phase space, which the authors argue is the basis of biological computation: a system's position in phase space functions as its memory, constraints, and structured dynamics. They propose an Observable Dimensionality Bound: external observers cannot track systems whose effective dimensionality exceeds a critical threshold set by channel capacity and temporal resolution, shielding internal structure from disruption. The framework suggests a thermodynamic pressure contributing to the efficiency gap between biological and artificial intelligence, as brains concentrate irreversible commitments at behavioral boundaries rather than every internal micro-update. Human cortex likely operates beyond this observable threshold.