Structural Balance in Signed Social Networks

Network Size

People in the network 9

Relationship Editor

Click one node, then a second node, to flip their relationship between friend and enemy.

Social Pressure Dynamics

Resolution speed 1×

Network State

Balanced triads

0 / 0

Under tension

0 edges

Rival factions

–

Repair steps

How it works

Every relationship (edge) carries a sign: +1 for friend, -1 for enemy. Heider's structural balance theory says a triad of three people is socially stable ("balanced") only if the product of its three edge signs is positive:

balanced(i,j,k)  ⇔  s(i,j) · s(j,k) · s(i,k) > 0

Stable patterns:  +,+,+  (three mutual friends)
                   +,-,-  (two friends share one enemy)
Unstable patterns: +,+,-  (a friend of my friend is my enemy — tension)
                    -,-,-  (three mutual enemies — tension)

An unbalanced triad is socially uncomfortable, so real networks drift toward balance: someone reconciles or a friendship breaks. Auto-resolve reproduces this by repeatedly finding an unbalanced triad and flipping the sign of one of its edges — the classic local-repair dynamics used to study opinion and alliance formation.

The Cartwright–Harary structure theorem proves that once a signed network is fully balanced, its nodes split into at most two mutually hostile factions: everyone inside a faction is a friend, everyone across factions is an enemy. This simulator computes that 2-faction split live and counts how many edges still contradict it ("under tension") — watch that number fall to zero as balance is reached.

Friend (+) Enemy (−) Faction A Faction B

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.