sim11: The Boundary Mechanism — Does Long-Range Inhibition Enable L2 Composition?
Two curvature-channel structures with a long-range inhibitor (far-field shadow − local material). Tests whether local activation + long-range inhibition (Turing/Gierer-Meinhardt) creates a boundary that prevents merging.
Step 0
A: Curvature + Inhibition, 2 seeds
—
B: Curvature + Inhibition, 1 seed (L1 control)
—
C: Curvature, no inhibition, 2 seeds (baseline)
—
D: Curvature, no inhibition, 1 seed (control)
—
Material Field (current snapshot)
Inhibitor Field (current snapshot)
Inhibitor not available for no-inhibition conditions.
Per-Region Components Over Time
Robustness Sweep (4 seeds)
inh_gain
seeds
l2_crossed
coexist
stable
h7
clean comp
Clean composition = 2-seed coexist AND 1-seed does NOT. The inhibitor converts 0/4 merge (no inhibition) to 2/4 clean coexistence at g=0.9 — a weak positive, not robust.