Recurrence Plot

A recurrence plot is a binary visualization of the times at which a dynamical system visits similar states in its phase space. Introduced by Eckmann, Kamphorst, and Ruelle in 1987, it is the foundational data structure from which recurrence networks are constructed. The plot is a square matrix where axes represent time, and each point (i, j) is marked if the state at time i is sufficiently close to the state at time j — revealing the deterministic structure hidden in apparently irregular dynamics.

The patterns visible in a recurrence plot — diagonal lines, vertical lines, checkerboard textures — are not aesthetic artifacts. They are signatures of dynamical properties. Long diagonal lines indicate that the system evolves through similar sequences of states, a signature of determinism. Short diagonal lines or isolated points indicate stochasticity or high-dimensional chaos. Vertical and horizontal lines indicate laminar states, where the system lingers in a particular region of phase space.

Recurrence plots have been applied to detect bifurcations, characterize synchronization between coupled systems, and distinguish chaos from noise — problems where traditional spectral methods fail because they assume linearity. The plot transforms a temporal signal into a spatial pattern, making the geometry of recurrence the primary object of analysis.

The recurrence plot is not merely a visualization tool. It is a radical reconceptualization of what it means for a system to be similar