Ornstein-Uhlenbeck process

The Ornstein-Uhlenbeck process is the mathematical model of a particle undergoing Brownian motion while subject to a linear restoring force — a random walk that is continuously pulled back toward a central tendency. It is the simplest stochastic process that exhibits both random fluctuation and mean reversion, making it the workhorse model for systems that wander but do not wander indefinitely: interest rates, neural membrane potentials, allele frequencies under mutation-selection balance, and the dominance durations of binocular rivalry.

The process is described by the stochastic differential equation:

dx = -θ(x - μ)dt + σdW

where μ is the long-term mean, θ is the rate of mean reversion, σ is the volatility, and dW is a Wiener process representing random shocks. The solution is a Gaussian process with autocorrelation that decays exponentially — the signature of a system with a single characteristic timescale.

The Ornstein-Uhlenbeck process is more than a statistical convenience. It is the stochastic analogue of a damped harmonic oscillator: where the deterministic oscillator settles into a fixed equilibrium, the OU process settles into a stationary distribution — a dynamic equilibrium that is never at rest but never diverges. This makes it the natural model for any system that is stable not because it resists perturbation but because it recovers from it.

In neuroscience, the OU process models the subthreshold membrane potential of a neuron, which fluctuates randomly due to synaptic bombardment but is restored toward rest by leak currents. In genetics, it models the drift of quantitative traits under stabilizing selection. In finance, it models interest rates that cannot wander too far from central bank targets. The same equation governs all three because the same structural pattern — random perturbation plus linear restoration — is substrate-independent.

The process connects to bistable systems in an interesting way. When an OU process drives a system with two stable states, the noise-induced transitions between states follow a distribution of waiting times that the OU autocorrelation structure can partially explain. The rivalry alternation durations, the switching times of genetic switches, and the regime changes in financial markets all share statistical signatures that may reflect a common underlying architecture: a bistable potential perturbed by correlated noise.