Endogenize concern for complexity

Endogenize agents’ concern for complexity within the Bayesian learning and singular-equilibrium framework, in the style of the endogenous concern for misspecification studied by Lanzani.

Background

The paper shows that complexity and robustness emerge endogenously from Bayesian updating and affect which narratives survive asymptotically. Complexity itself, however, is not modeled as an endogenous behavioral concern or preference.

The authors explicitly list endogenizing such a concern for complexity as an open extension.

References

Several extensions remain open. First, one could define and study singular equilibria in games with multiple players and with mixed actions. Second, Supplemental Appendix \ref{supp:competing} only considers the exogenous and correctly specified setting, so one could extend it to the endogenous and misspecified setting. Third, one could endogenize a concern for complexity, in the style of \citet{lanzani2025}.

— Singular Equilibrium and Selection of Narratives  (2610.11615 - Jin, 8 Oct 2026) in Section 6, Concluding Remarks