Message Order Selects Opposite Collective Opinions on Laplacian-Cospectral Networks

2026-08-20

AI Summary

研究通过构建量子类网络模型，探讨了消息顺序对整个网络最终意见的影响，揭示了拉普拉斯同谱网络中不同意见形成的机制。

Abstract

Receiving the same messages in different orders can change an agent's judgement. We construct a network model in which this local sequence effect determines the final opinion of an entire network. Each agent has a two-level quantum-like state with an expressed opinion and an auxiliary context coordinate that can affect the next update. Two valid update rules of equal strength leave the neutral state unchanged and add no positive or negative opinion bias. Nevertheless, reversing their order gives the exact opinions +0.01 and -0.01 from the same initial state. We place four ordered pairs so that their positive and negative contributions sum to zero but occupy different labelled nodes. After this single input, all cases follow the same nonlinear network rule, which preserves valid states and admits two stable consensuses. We compare two 16-node networks whose complete Laplacian spectra are identical. With every local ingredient fixed, they converge to opposite consensuses. A separate local calculation identifies a specific overlap between the labelled pulse and Laplacian eigenspace projectors that sets the leading displacement of the basin boundary. Rigorous interval arithmetic proves the finite registered example independently. The construction connects message order to collective selection and shows why Laplacian eigenvalues alone do not determine which opinion wins.

Problem

• Message Order
• Laplacian-Cospectral Networks
• Collective Opinions

Innovation

• message order
• Laplacian-cospectral networks
• quantum-like state
• collective selection
• opinion consensus