When Can Decision Reasoning Cross Domains?: Testing Field-Based and Conventional Methods Under Changing Conditions
Year: 2026
Type: preprint
Abstract: Complex decision systems depend on evidence generated under heterogeneous models, objectives, operational constraints, and changing environments. A procedure that helps revise a decision in one domain may be reusable elsewhere, but neither the validity of its evidence nor the authority to act necessarily travels with it. Established methods in causal transportability, adaptive planning, provenance, model credibility, safety assurance, and evidence-based practice already address important components of this problem. This conceptual paper asks when a common decision-qualification method may legitimately cross domain boundaries, and whether a mission-conditioned Field representation contributes methodological value beyond an information- and resource-matched conventional alternative. Six constructed cases are organized into three deliberately heterogeneous contrasts: healthcare and scientific discovery; electrical power and autonomous robotics; and climate adaptation and logistics. Each case differentiates the claim under examination, the domain-specific warrant required, the change event, and the conditions for warranted revision, rejection, or indeterminacy. A provisional Cross-Domain Correspondence Gate distinguishes reuse of an analytical question, structural correspondence, and transport of substantive scientific conclusions, while a three-level adjudication separates claim validity, comparison admissibility, and net methodological value. Strong baselines integrate disciplinary models with assurance cases, provenance, adaptive decision methods, and independent review. Adversarial controls test false analogies, unsupported strengthening of claims, stable recommendations with changing justifications, conventional equivalence, and unnecessary representation overhead. The study offers a bounded comparative research protocol, not a new causal identification theorem, universal ontology, empirical benchmark, or demonstrated Field advantage. Its outcomes may ultimately include limited utility, conventional sufficiency, negative net value, invalid transfer, or inconclusive evidence. A prospective validation agenda specifies what independent testing would be needed to discriminate among these possibilities.
Author: Figurelli, Rogério
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7236824692
