The rational agent model does not fail because humans are irrational. It fails because it assumes coherence across decision domains that does not exist. This essay starts with a concrete case — a practicing Catholic who visits a brothel every Saturday night and returns to his family Sunday morning without apparent conflict — and derives a structural claim: the same agent operates under incompatible systems in parallel, without transition and without residue. There is no unified preference function. The inference is structurally blocked. The implications reach beyond psychology into model design, institutional architecture, and risk. Systems built on cross-domain coherence do not degrade gradually. They collapse at the moment of domain shift — exactly when they need to hold.
Building similarity graph...
Analyzing shared references across papers
Loading...
Javier Ignacio Janer Tittarelli
Building similarity graph...
Analyzing shared references across papers
Loading...
Javier Ignacio Janer Tittarelli (Wed,) studied this question.
www.synapsesocial.com/papers/69d896046c1944d70ce07393 — DOI: https://doi.org/10.5281/zenodo.19474566