Transient disorder in dynamically growing networks

When a certain "seed" disturbance begins to spread on a large network, the number of nodes infected is a function of time. Regarding the set of infected nodes as constituting a dynamic network that evolves continuously in time, we ask: how does the order in the collective dynamics of the network vary with time? Utilizing synchronizability as a measure of the order, we find that there exists a time at which a maximum amount of disorder corresponding to a minimum degree of synchronizability can arise before the system settles into a more ordered steady state. This phenomenon of transient disorder occurs for networks of both regular and complex topologies. We present physical analyses and numerical support to establish the generality of the phenomenon.