Identity as a First-In Cache: Anti-Aliasing, Monotone Fictionalization, and the Coherence Kernel

Humans misperceive autobiographical memory as a long, uniformly detailed film. In practice, it behaves like a rolling cache: a small working set of recent, high-resolution states plus heavily compressed summaries of the past. We formalize this as a First-in Cache Identity (FCI) model in which perception is compressed by a channel C, maintained in a finite working set Wt, anti-aliased by an operator Bthat hides cache rotation, and recalled by Rto produce a coherent narrative. Two invariants drop out immediately. (1) Monotone fictionalization: Fisher information (phenomenal curvature) of recalled episodes decreases monotonically with age under memory channels independent of the latent parameters. (2) Lyapunov coherence: under generic forgetting flows the Fisher metric contracts, making continuity a stable fixed point maintained by a coherence kernel (the ego). We connect this psychology-scale account to the broader Abzu corpus: QFI-based compiled law, sheaf-theoretic context gluing, and the QFI →TC bridge for phase transitions in cognition. We propose experiments in humans and artificial agents to measure curvature-vs-age, blur gain vs. confabulation, and phase switches as cache size or compression rate varies.