🔬 condensed matter

Persistence, resetting, and first-passage times of an active Ornstein--Uhlenbeck particle

This paper analytically investigates how stochastic resetting and active persistence jointly influence the first-passage time statistics of an active Ornstein--Uhlenbeck particle on a finite interval, deriving conditions under which these mechanisms synergistically reduce the mean first-passage time compared to passive diffusion.

Original authors: Demosthenes K. Georgiou, Paul C. Bressloff, Thibault Bertrand

Published 2026-09-21

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Demosthenes K. Georgiou, Paul C. Bressloff, Thibault Bertrand

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the natural world, finding a target is often a matter of time and chance. Whether it is a bacterium hunting for nutrients, a protein folding into its functional shape, or an animal searching for food, the process is fundamentally about how long it takes a wandering entity to reach a specific destination for the first time. Scientists call this the first-passage time. For simple particles moving randomly, like dust motes in a sunbeam, this journey is governed by diffusion, a slow, jittery drift. However, many living systems are not passive; they consume energy to move with purpose. These are active particles, such as swimming bacteria or synthetic micro-robots, which propel themselves forward with a degree of persistence, maintaining their direction for a short while before tumbling or changing course. Understanding how this self-driven motion affects search times is crucial, but the picture becomes even more complex when we introduce the concept of resetting. Resetting is a mechanism where a system is randomly interrupted and returned to a starting point, a strategy that can sometimes turn a hopeless, endless search into a successful one. The question remains: how does the combination of self-propelled motion and these random restarts alter the time it takes to find a target?

A team of researchers at Imperial College London has tackled this question by studying a mathematical model of an active particle confined to a narrow, one-dimensional corridor. Imagine a tiny swimmer trapped between two walls, with one wall acting as a trap that ends the journey and the other acting as a mirror that bounces the swimmer back. This swimmer is an active Ornstein–Uhlenbeck particle, a theoretical construct that moves with a velocity that fluctuates randomly but tends to stay correlated over a short period, mimicking the persistence of real biological swimmers. The researchers wanted to know how long it takes this swimmer to hit the absorbing wall, and how that time changes if the swimmer is occasionally snatched back to its starting position and given a fresh, random push. They focused on a regime of weak activity, where the self-propulsion is present but not overwhelmingly strong, allowing them to use a step-by-step mathematical approach to solve the complex equations governing the particle's motion.

The study revealed that the effect of the particle's own activity on its search time is not straightforward; it can either help or hinder the journey, depending entirely on the relationship between how long the particle keeps its direction and how long it takes to drift to the target. If the particle starts very close to the absorbing wall, its self-propulsion can actually slow it down. This happens because the particle might be pushed away from the target, traveling deep into the middle of the corridor before its direction changes, effectively wasting time. However, if the particle starts further away, or if its direction changes very quickly, the extra energy from self-propulsion acts like a boost, helping it reach the target faster than a passive, drifting particle would. The researchers found a specific crossover point where the benefit of activity flips to a cost, determined by the ratio of the particle's persistence length to the size of the container.

When the researchers introduced the element of resetting, they discovered that this intervention could rescue the search process, but only under specific conditions. If the particle starts too close to the target, resetting is counterproductive; it interrupts trajectories that were already on a fast track to success, effectively resetting the clock on a journey that was about to finish. In these cases, the more often the particle is reset, the longer it takes to arrive. However, if the particle starts further away, resetting becomes a powerful tool. By periodically returning the particle to the start, the system prevents it from getting lost in long, unproductive excursions into the middle of the corridor. The researchers identified a sweet spot, an optimal rate of resetting, that minimizes the total search time. This optimal rate exists only when the particle is far enough from the target that the risk of getting lost outweighs the risk of interrupting a good run.

A particularly subtle finding emerged when the researchers considered how the particle's velocity is handled during a reset. In their model, a reset does not just return the particle to its starting position; it also reassigns a new propulsion velocity. The researchers found that the outcome of this process depends heavily on the direction of that new velocity. If the particle is consistently reset with a velocity pointing toward the target, the search becomes significantly more efficient, and resetting remains beneficial even for particles starting quite far away. Conversely, if the reset velocity points away from the target, the benefits of resetting are diminished, and the range of starting positions where resetting helps shrinks. This highlights that the mechanics of the reset itself—what state the system is returned to—are just as important as the frequency of the reset.

Finally, the team constructed a comprehensive map of the conditions under which an active, resetting particle outperforms a simple, passive one. They found that the competition between three distinct timescales dictates the outcome: the time it takes to diffuse to the target, the time the particle maintains its direction, and the average time between resets. When the particle's persistence is short, or when the resetting is frequent enough to keep the particle from wandering too far, the active system with resetting can find the target faster than a passive system ever could. However, if the particle is too persistent and the resetting is too infrequent, the active particle gets stuck in its own momentum, and the passive, drifting particle wins the race. The study concludes that there is no single rule for the best search strategy; instead, the most efficient approach depends on a delicate balance between how long the searcher holds its course, how far it must travel, and how often it is forced to start over.

Technical Summary: Persistence, Resetting, and First-Passage Times of an Active Ornstein–Uhlenbeck Particle

Problem Statement

This study investigates the first-passage time (FPT) statistics of an active Ornstein–Uhlenbeck particle (AOUP) confined to a finite one-dimensional interval $[0, L]$ . The system features an absorbing boundary at $x=0$ and a reflecting boundary at $x=L$ . The particle is subject to stochastic resetting, where the system is randomly interrupted and returned to a predetermined state.

The core challenge addressed is the combined effect of activity (self-propulsion with finite persistence time) and stochastic resetting on the mean first-passage time (MFPT). While FPT statistics for passive Brownian particles under resetting are well-established, the interplay between activity and resetting in active matter models remains less understood. Specifically, the paper examines how the resetting of both the kinematic variable (position) and the dynamic variable (propulsion velocity) influences search efficiency and the conditions under which resetting accelerates or delays absorption.

Methodology

The authors employ a combination of analytical perturbation theory and numerical simulations:

Model Formulation: The AOUP dynamics are governed by coupled Langevin equations for position $x(t)$ and self-propulsion velocity $v(t)$ , where $v(t)$ follows an Ornstein–Uhlenbeck process characterized by a persistence time $\tau_p$ and noise strength $D_v$ .
Perturbative Analysis: Working in the weak-activity regime (small Péclet number, $Pe \ll 1$ $P e ≪ 1$ ), the authors develop a regular perturbation expansion of the Fokker–Planck equation. The expansion parameter is $\epsilon = \sqrt{Pe}$ $ϵ = P e$ .
- The joint probability density is expanded in powers of $\epsilon$ .
- The velocity dependence is diagonalized using Hermite polynomials, reducing the problem to a set of coupled ordinary differential equations for the spatial coefficients.
- Analytical expressions are derived for the Laplace-transformed survival probability $\hat{S}_0(s)$ and the FPT distribution moments up to order $O(Pe^2)$ .
Renewal Framework: To incorporate Poissonian resetting at rate $r$ , the authors utilize a renewal equation. This framework relates the survival probability with resetting, $\hat{S}_r(s)$ , to the survival probability without resetting, $\hat{S}_0(s)$ , via the relation $\hat{S}_r(s) = \hat{S}_0(s+r) / [1 - r\hat{S}_0(s+r)]$ .
Validation: The analytical results are validated against direct numerical simulations. The perturbation theory is shown to remain accurate up to $Pe \approx 1$ .
Velocity Resetting Protocols: The study compares two protocols:
- Stationary Gaussian: The velocity is redrawn from the stationary Gaussian distribution $\rho_0(v)$ upon resetting.
- Fixed Velocity: The velocity is reset to a fixed value $v_0 = \pm 1$ . (Results for this case are obtained via simulation as the perturbative framework assumes a stationary Gaussian distribution).

Key Contributions and Results

1. Effect of Activity without Resetting

In the absence of resetting, activity can either increase or decrease the MFPT depending on the ratio of the persistence time ( $\tau_p$ ) to the diffusive absorption time ( $t_d \sim x_0^2/D$ ).

Short Persistence ( $t_d \gg \tau_p$ ): The propulsion velocity decorrelates before absorption. The motion becomes effectively diffusive with an enhanced diffusivity, reducing the MFPT.
Long Persistence ( $t_d \ll \tau_p$ ): The particle maintains its initial velocity. If directed away from the target, it travels deep into the bulk, increasing the MFPT.
Crossover: A critical initial position $x_c$ exists where the effect of activity switches from beneficial to detrimental. This crossover depends on the scaled persistence length $\lambda/L$ .

2. Onset of Beneficial Resetting

The introduction of resetting leads to two distinct regimes based on the initial position $x_0$ :

Non-monotonic Regime: For particles starting sufficiently close to the absorbing boundary ( $x_0 < x^*_0$ ), the MFPT $T_r$ exhibits a minimum at an optimal resetting rate $r^*$ . Resetting suppresses long excursions into the bulk without frequently interrupting trajectories that would otherwise reach the target.
Monotonic Regime: For particles starting further away ( $x_0 > x^*_0$ ), resetting increases the MFPT monotonically by repeatedly returning the particle to a position far from the target.
Shift in Critical Position: The critical position $x^*_0$ separating these regimes shifts to larger values (further from the boundary) as activity ( $Pe$ ) increases. This implies that resetting remains beneficial for a wider range of starting positions in active systems compared to passive ones.

3. Range of Beneficial Resetting Rates

Even when resetting is initially beneficial (i.e., $dT_r/dr|_{r=0} < 0$ ), there exists a finite range of resetting rates $r < r_c$ for which the MFPT is lower than the no-resetting active MFPT. Beyond $r_c$ , resetting becomes detrimental. The range of beneficial rates shrinks as $x_0$ approaches the critical position $x^*_0$ .

4. Influence of Velocity Resetting Protocols

The nature of the velocity reset significantly alters the phase boundaries:

Resetting to a fixed velocity directed towards the target ( $v_0 = -1$ ) shifts the beneficial regime to larger initial positions.
Resetting to a fixed velocity directed away from the target ( $v_0 = +1$ ) shifts the beneficial regime to smaller initial positions.
These shifts scale approximately linearly with $\sqrt{Pe}$ .

5. Active vs. Passive Search Phase Diagram

By comparing the active particle with resetting to a passive Brownian particle without resetting, the authors construct a phase diagram in the $(x_0/L, \tau_p)$ plane.

Blue Region ( $T_{rel} < 1$ ): Activity and resetting combined reduce the MFPT compared to passive diffusion without resetting.
Red Region ( $T_{rel} > 1$ ): The combined process is slower than passive diffusion.
The boundary is determined by the competition between three timescales: diffusive absorption time, persistence time, and mean reset time. For small $x_0$ , the active-resetting process is generally favored. For large $x_0$ , the passive process becomes more efficient as persistence increases, unless the reset rate is high enough to suppress persistence effects.

Significance and Claims

The paper claims to provide the first analytical treatment of FPT statistics for an AOUP under stochastic resetting in a confined geometry. Its primary significance lies in:

Analytical Framework: Extending perturbative methods (previously used for steady-state behavior) to first-passage problems with resetting, yielding closed-form expressions for FPT distributions and moments.
Mechanistic Insight: Demonstrating that the interplay between activity and resetting is governed by the competition between persistence and resetting timescales.
Protocol Dependence: Highlighting that the efficacy of resetting in active systems is not universal but depends critically on how the propulsion velocity is reinitialized.
Optimization: Identifying specific regions in parameter space where the combination of activity and resetting outperforms passive diffusion, offering a theoretical basis for optimizing search strategies in active matter systems.

The authors note that their results are restricted to the weak-activity regime and that future work is needed to explore strongly persistent regimes and alternative reset protocols (e.g., fixed velocity) analytically. They also suggest that extending the model to partially absorbing targets would be relevant for biological search processes.