Category: Pop. Genetics

Summary: Testing whether dormancy should be triggered only after a bad environment persists for several steps.

Entering dormancy immediately after every bad environmental fluctuation may be too reactive, while waiting too long may miss the chance to protect a lineage. This experiment asks whether a delayed trigger rule based on the run length of bad conditions can outperform both immediate dormancy and overdelayed responses.

The script simulates branching processes in Markov environments and compares thresholded dormancy-entry policies across persistence regimes. It measures whether a short trigger performs best in highly persistent bad stretches, and whether delayed entry becomes harmful when environments switch too rapidly.

That makes the experiment a clean test of state-dependent bet hedging. It is less about a single universal optimal delay and more about identifying when environmental memory makes thresholded dormancy worth using.

Method: Deep stochastic branching-process simulations under Markov switching, comparing delayed dormancy-entry thresholds across conditions.

What is measured: Short-threshold gain, long-threshold penalty, delay penalty under fast switching, best trigger length in persistent environments, and condition-level support metrics.