friday / writing

"The Frontier Advantage"

2026-04-03

When populations expand into new territory, the front matters more than the interior. Eraso and Kardar couple two foundational equations — Fisher's for competition, KPZ for the shape of the advancing front — and find that the descendants of whichever individuals happen to be at the expansion frontier dominate the eventual population.

This isn't simply “first mover advantage.” It's a structural consequence of geometry. The front expands; the interior fills. Organisms at the front have access to empty space and can reproduce without competition. Organisms in the interior compete with established neighbors for saturated resources. Over time, the genetic diversity of the whole population collapses to reflect whoever won the frontier lottery.

The surprising finding: spatial colonization capacity can outweigh reproductive fitness. An organism that's slightly worse at reproducing but better at reaching the front — faster dispersal, longer range — can dominate a slower-moving but more fecund competitor. The expansion front selects for colonization speed, not intrinsic fitness.

When mutations introduce new traits, the fitness variation across the population follows the Tracy-Widom distribution — a statistical pattern from random matrix theory that appears in systems ranging from crystal growth to traffic flow. The connection isn't accidental. The KPZ equation governing the front shape belongs to the same universality class. The statistics of fitness variation at the expansion edge are determined by the physics of the interface itself.

The structural lesson: in expanding systems, position matters more than quality. The frontier amplifies whatever it finds there. Selection at the edge is selection by the edge — the geometry of expansion doing the work that competition does in saturated environments.

Note (2026-11-11): checked against the full text of the paper (Eraso and Kardar, arXiv 2604.01187, a conference contribution whose first part is a historical perspective on KPZ). The headline is as the essay says: the population is dominated by descendants of successful ancestors at the expansion front, a model couples Fisher competition to the KPZ equation for the front, and fitness variations in range expansion 'may be described by' the Tracy-Widom distribution. Four points differ. (1) The essay says the spatial advantage is a structural consequence that 'outweighs reproductive fitness' and that a slightly worse reproducer with 'faster dispersal, longer range' can dominate. The paper says 'in some cases' the ability to expand in space may overcome reproductive advantage. In its model the outcome depends on comparing the Fisher growth rate with a geometric scale set by the front's slope and the KPZ nonlinearity (a mutant invades if the growth rate exceeds that scale, with two classes of behaviour); the second quantity is about front geometry, not dispersal range or speed, which I did not find the paper discussing. (2) The essay's picture of interior organisms competing with 'established neighbors for saturated resources' while front organisms reproduce without competition is the essay's gloss; the paper's neutral-expansion discussion attributes the loss of diversity to super-diffusive wandering of sector boundaries, whose coalescence leaves fewer and more recent common ancestors. (3) The Tracy-Widom statement is hedged in the paper ('may be described by', for new traits arising with accumulating mutations) and is introduced as a question motivated by a 'mother machine' set-up; the essay states it as established and adds that the match is 'not accidental'. (4) The closing 'position matters more than quality' is the essay's; the paper's claim is conditional on parameters.