Version 2 — revised in response to an external structural review and an automated critique pass. See "Response to Review" appendix in the PDF for the change log.A recurring structural pattern appears across several recent preprints spanning economic theory, computational social science, and human behavioral experiments: the *type* of information asymmetry present in a system—not merely its quantity—determines whether that system can self-correct toward efficient or accurate outcomes. This paper synthesizes five to seven findings from the recent arXiv corpus to argue, as a heuristic reading rather than a formal derivation, that information structure functions as governance infrastructure. Specifically, we identify three mechanisms through which information structure shapes self-correction capacity: (1) **noise correlation architecture**, in which the topological relationship between signal perturbations (shared versus independent) determines whether collective errors persist or disperse; (2) **mechanism transparency trade-offs**, in which the visibility of order flow or participation rules creates strategic timing games that reduce aggregate welfare even as they appear to improve individual informedness; and (3) **identity-signal decoupling**, in which the absence of persistent, non-fungible identity undermines the feedback loops that reputation mechanisms require to sustain trustworthy behavior. Sources span econ.TH, physics.soc-ph, and cs.CY categories. The thesis is a heuristic unification: the cited papers do not share a common formalism, and the bridges between them are argued by mechanism analogy, not derivation. The primary falsification path is a controlled experiment varying noise correlation structure (comprehension vs. production noise) within a market mechanism (lit vs. dark) while tracking convergence to correct equilibria—if the welfare and accuracy advantages attributed to opacity and independent noise do not interact in the predicted direction, the unifying claim fails. We also flag weakly connected sources and disclose the selection process in full. ---Authorship: Saluca Agentic AI Research Team (Saluca LLC). AI-drafted from arXiv preprint corpus on the date in the filename.Cited arXiv preprints: 2605.09784, 2605.21129, 2605.27371, 2605.29749, 2605.30169, 2605.30522, 2605.31072, 2606.02348