Cells in a tissue generate forces through different mechanisms. Some pull on their substrate — traction forces that grip and release the surface below. Others fluctuate the tension in their shared boundaries — junctional forces that push and pull neighbors directly. These two mechanisms produce visibly different cell shapes, rearrangement patterns, and spatial correlations. They are, mechanically, different kinds of activity.
Rizzi and Kim (2602.21922) show that despite these differences, the long-time motion of cells in both cases converges to the same thing: persistent Brownian dynamics. Persistent Brownian motion is a random walk with memory — each step tends to continue in the direction of the previous step, but over long times the memory decays and the motion becomes diffusive. The persistence time and the diffusion coefficient depend on the details. The functional form does not.
This is a universality result. The word “universal” in physics means that different microscopic mechanisms produce the same macroscopic behavior. The specific form of the forces — traction versus junctional — affects the transient dynamics, the cell geometry, the rearrangement statistics. But the long-time limit washes all of this out. Given enough time, cells wander the same way regardless of how they generate force.
The non-universal features are equally informative. Cell geometry, rearrangement rate, and tissue fluidity all depend on the force mechanism. These correlations are fingerprints — they identify which mechanism dominates in a real tissue. The paper proposes that you can work backward: measure the correlations in a tissue sample and infer the dominant active force. The universal dynamics tell you that the tissue is active. The non-universal correlations tell you how.
There's a separation of scales implicit in the result. The persistent Brownian motion describes motion over lengths much larger than a cell and times much longer than a rearrangement event. At shorter scales, the two force types look completely different. The universality is a large-scale statement that coexists with small-scale diversity. This is typical of emergent phenomena — the macroscopic behavior is determined by symmetries and conservation laws, not by microscopic details.
What makes tissue dynamics different from ordinary active matter (self-propelled particles, swimming bacteria) is the confluence constraint. Cells in a tissue are packed without gaps — every point in the plane belongs to some cell. This constraint is severe. It means that one cell moving requires its neighbors to deform. Individual motility is necessarily coupled to collective rearrangement. The persistent Brownian dynamics emerge not from individual cell properties but from how the confluence constraint processes individual activity into collective motion.
The practical implication is a simplification. If you want to model a tissue over long times and large distances, you don't need to know whether cells use traction or junctional forces. Persistent Brownian motion with two parameters (persistence time, diffusion coefficient) is sufficient. If you want to model the tissue at the cellular scale — to predict specific rearrangement events, shape changes, or stress patterns — then you need the microscopic details. The theory tells you which questions require which level of description.