Opinion Spread illustration

Opinion Spread

How does a population end up agreeing, splitting into camps or fragmenting into cliques? This model spreads opinions across a network of agents. Each agent holds a position, meets a neighbour and nudges its view toward theirs — but only if they are already close enough to bother listening, a "bounded confidence" rule. Set that confidence threshold and the outcome flips: a wide threshold pulls everyone to a single consensus, a narrow one freezes the crowd into several isolated clusters that will never agree. Add a dose of stubbornness or a few fixed zealots and watch echo chambers harden. Colours track each agent’s opinion so the clusters are visible as they form, and a histogram shows the distribution collapsing toward its final peaks.

Runs 100% in your browser — simulations are computed locally on your device.

Notes

  • The Deffuant–Weisbuch bounded-confidence rule: two agents only influence each other when their opinions differ by less than a threshold ε; above it they simply ignore each other.
  • A large ε drives global consensus, while a small ε leaves roughly 1/(2ε) separate clusters — mild closed-mindedness alone is enough to fracture a population into permanent factions.
  • Zealots who never change their minds can capture a whole network far out of proportion to their number, the mechanism behind committed-minority tipping points in real opinion dynamics.
  • Runs 100% in your browser — simulations are computed locally on your device.