How dense must the contact network be before an outbreak reaches most of the population? At the default link-chance of 8, contact-count is about 30 and most people are unlinked dots, so an outbreak can only reach the cluster it starts in.
Try this Set link-chance to 40 and press Reset (it is read only at setup). About 100 of the 120 people end up in one connected cluster, and the outbreak typically reaches 90 or more of them.
About this model Where it ends up
An SIR epidemic on a contact network, after Stonedahl and Wilensky's NetLogo Virus on a Network (2008). The virus can only travel along links, and at the default settings most people have none, so an outbreak rarely gets beyond the cluster it starts in.
Setup scatters population people at random and infects the first initial-outbreak-size. Each person has a link-chance percent chance of linking to 3 random others; position plays no part, and a force layout then pulls linked people together. Each tick, in random order, a susceptible person with any infected contact is infected at virus-spread-chance percent (one try, however many contacts are infected), and an infected person recovers permanently at recovery-chance percent. Updates are sequential, so the virus can cross more than one link in a tick. The two chances are read every tick; the other params apply at Reset.
Differences from the NetLogo model: there, each node links to its nearest neighbors in space until an average degree is reached, each infected neighbor gets its own chance to transmit, and recovered nodes become susceptible again unless they gain resistance. Here links ignore position, transmission is one try per susceptible person, and recovery is permanent.
