Arc of Self: A Recursive Theory of Identity

Identity is modeled as a fourteen-node re-entrant architecture in four phases (Hardware, OS Install, Engine, Loop), with a closed running circuit at nodes 10-14. Cultural concepts, installed through an innate loader, assemble into Personality Properties; the Properties generate a standing demand for validation; the demand shapes attention and situation selection; behavior is executed to confirm the self-model; and the resulting feedback is filtered in the self-model's favor before it can revise the concepts that produced it. The framework synthesizes established results (behavioral genetics, core knowledge, predictive processing, self-verification, dissonance, hedonic adaptation); its contributions are the ordered architecture itself, the full running pass as the unit of analysis, an account of emotional concepts as full-loop configurations with addressable hot nodes, and the Collision Layer, a mechanistic description of how two identity loops modify each other through unlabeled behavioral output. Six falsifiable predictions are staked. The interactive form of the theory is available at https://arc-of-self.com.