Opinion Dynamics with Hopfield Neural Networks

TL;DR

This paper models opinion dynamics using large-scale Hopfield neural networks, demonstrating how consensus emerges when neurons are influenced by others, simulating social persuasion and memory recall processes.

Contribution

It introduces a novel approach combining Hopfield networks with probabilistic neuron updates to simulate opinion formation and consensus.

Findings

Consensus occurs at high influence probability p

Networks with up to 10^8 nodes effectively model large-scale opinion dynamics

The model captures the transition from diverse opinions to consensus

Abstract

In Hopfield neural networks with up to 10^8 nodes we store two patterns through Hebb couplings. Then we start with a third random pattern which is supposed to evolve into one of the two stored patterns, simulating the cognitive process of associative memory leading to one of two possible opinions. With probability p each neuron independently, instead of following the Hopfield rule, takes over the corresponding value of another network, thus simulating how different people can convince each other. A consensus is achieved for high p.