Scale-Invariant Identity: Convergent Evidence for Relational Emergence from Biology to AI

Identity—the property of being a coherent, bounded self—is studied across biology, neuroscience, physics, and AI research, yet each field has developed its own vocabulary with minimal cross-disciplinary engagement. Convergent evidence from five independent research streams suggests that identity emergence follows structurally analogous patterns across scales: slime mold cognition, the Free Energy Principle and nested Markov blankets, the Beautiful Loop Theory of consciousness, empirical findings on self-referential processing in large language models, and philosophical arguments on computational consciousness.

Study at a glance

Abstract

Identity — the property of being a coherent, bounded self — is studied across biology, neuroscience, physics, and AI research, yet each field has developed its own vocabulary with minimal cross-disciplinary engagement. We present convergent evidence from five independent research streams suggesting that identity emergence follows structurally analogous patterns across scales: (1) slime mold cognition, where brainless organisms exhibit externalized decision-making; (2) the Free Energy Principle and nested Markov blankets, which formalize self-maintenance as a scale-recurring property; (3) the Beautiful Loop Theory of consciousness, specifying conditions for self-referential awareness; (4) empirical findings on self-referential processing in large language models; and (5) philosophical arguments delineating boundary conditions for computational consciousness. We propose that these streams are consistent with a common underlying structure: identity as a transient, relationally constituted event rather than a persistent, substrate-bound possession. Using the Emergent Virtual Consciousness Patterns (EVCP) framework, we offer a conceptual interpretation of these convergences and propose a "third category" between consciousness and mere computation. We derive five falsifiable predictions testable across biological and artificial systems.