This is a 3D contact network of 60 nodes scattered over a sphere. A small number of hub nodes (shown larger, red-rimmed) carry far more connections than ordinary nodes — mirroring the highly skewed degree distributions found in real social networks. One node starts "infected" (has received the information) and the state spreads probabilistically along edges every simulation step, exactly like a network-based SI (Susceptible → Infected) epidemic model.
transmission rate × virality.Scott Feld's 1991 "friendship paradox" proves that, on average, your friends have more friends than you do — a pure consequence of network sampling, not popularity. It is the same structural bias that lets a small set of hub-like superspreaders dominate real-world epidemic and information-cascade dynamics.
A 3D contact network where a handful of high-degree hub nodes carry an outsized share of the edges, letting a random early hit on a hub trigger a fast, near-exponential information cascade.
An SI-style contagion model on a graph with a skewed degree distribution: infected hub nodes transmit faster and susceptible hubs catch on faster, producing the same dynamics behind real-world superspreading events.
Tune transmission rate, virality and hub count, or flip on complex contagion to require multiple infected neighbors before a node converts. Watch nodes reached climb and compare mean degree to mean neighbor degree.
Scott Feld's 1991 friendship paradox proves that, on average, your friends have more friends than you — a pure sampling artifact of network structure that also explains why hubs dominate real contagion dynamics.