Rebellions and Impeachments in a Neural Network Society

TL;DR

This paper models political dynamics in South American democracies using neural networks, revealing how legislative focus and presidential approval influence impeachment likelihood.

Contribution

It introduces a novel neural network-based statistical mechanics model to analyze collective political actions leading to impeachment.

Findings

Increased presidential agenda items reduce cross-party dialogue.

Lower presidential approval correlates with higher impeachment risk.

The model predicts impeachment likelihood based on political variables.

Abstract

Basede on a study of the modern presidencial democracies in South America, we present a statistical mechanics exploration of the collective, coordinated action of political actors in the legislative chamber that may result on the impeachment of the executive. By representing the legislative political actors with neurla networks, we observed that the larger the effective number of presidential-agenda items are treated, the smaller the chances for a cross-party dialogue, which, if combined with a decrement in the president's public approval rating, could trigger an impeachment process.