Information Acquisition Towards Unanimous Consent

2024-05-28

AI Summary

This paper addresses mechanism design under unknown task difficulty, where a principal must elicit information from privately informed agents (workers) via low-cost tests while ensuring incentive compatibility and consensus. Challenges include strategic misreporting of test outcomes and strong type privacy. Method: The paper establishes that threshold testing is optimal for information acquisition; it introduces, for the first time, a multi-interval two-stage testing mechanism that guarantees existence of a Bayesian equilibrium and achieves consensus even under manipulable information structures. It jointly optimizes mechanism design and information architecture. Contribution/Results: The framework significantly improves the principal’s expected payoff and expands the feasible consensus region. Its core contribution lies in departing from conventional single-point testing paradigms to propose a dynamic screening framework that jointly ensures information efficiency and incentive compatibility.

Technology Category

• Game Theory and Economic Paradigms: Mechanism Design
• Multiagent Systems: Mechanism Design
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

A manager facing a task of unknown difficulty can propose a plan to let a worker undertake the task; the worker can either accept the proposal or reject it. The plan benefits the worker only when the task is sufficiently easy and benefits the manager only when it is sufficiently hard. The manager can conduct a test at no cost to acquire information about the difficulty of the task; however, she can misreport the test result to the worker. We find that it is optimal for the manager to conduct a threshold test and to propose the plan only when the difficulty of the task exceeds the threshold. Moreover, when the worker privately knows his capability, we find that the manager can benefit from screening the worker by offering up to two additional interval tests.

Problem

• Optimal test strategy for task difficulty assessment
• Manager-worker proposal under information asymmetry
• Screening worker capability with additional interval tests

Innovation

• Threshold test for task difficulty assessment
• Optimal plan proposal based on threshold
• Screening workers with interval tests