no. 14 · Friday 13 November 2026
Round-Based Contagion Generates Chaos from Oscillatory Dynamics: A Two-Stage Contagion Map
Abstract
Many social processes unfold in rounds, such as elections, meetings or turns of a game, in which individuals update synchronously. We show that this round structure can turn a regular oscillation into chaos. In Katriel's differential equations, two-stage contagion (adoption after two contacts with adopters) combined with population turnover produces bistability and relaxation oscillations, but no chaos. From the elementary reactions we derive a deterministic map $F_τ$ for a large population updated in rounds of length $τ$: each individual makes at most one transition per round, events compete as exponential risks, and adoptions take effect from the next round. Katriel's equations are recovered as $τ\to0$. For short rounds the map follows the relaxation cycle, and the phase of successive outbreaks obeys a circle homeomorphism, which excludes chaos. As the rounds lengthen, this circle map loses invertibility and the invariant curve breaks down; once a round lasts about two mean adoption times, comparable to the duration of an outbreak, the attractor is chaotic, while Katriel's equations remain on a periodic relaxation cycle. With the probability $p$ of adoption after a single contact as bifurcation parameter, the map goes from extinction through a window of chaos to a stable equilibrium; for $β/γ=46$, $μ/γ=0.002$, $γτ=5.5$ the window is $0.023\le p\le0.056$, bounded by the invasion threshold and by a boundary crisis, whereas Katriel's equations oscillate for $0.022<p<0.41$. Chaos is supported by the Lyapunov spectrum (Kaplan–Yorke dimension $1.81$), the correlation dimension, the 0-1 test, ordinal-pattern complexity, saddle periodic orbits and a horseshoe of the phase map. One-stage contagion, $p=0$ and permanent adoption showed no chaos.
- comments
- 29 pages, 12 figures, 8 tables; includes Supplementary Material (7 pages, 5 figures, 3 tables)
- fields
- Dynamical Systems