When does a developing creature become harder to redirect?
At nine times during each organism’s development, we forced the reservoir to supply founder 0, 1, 2, or 3—one event only—then released the system and watched twelve future passages. The spread among those four futures is the causal susceptibility of that state.
plastic
still open
closing edge
narrower
still narrow
causal susceptibility = mean pairwise distance among four forced-founder futures
A plateau, then a step.
Morphological leverage is essentially unchanged from two passages before the information turn through the turn itself. Then it falls sharply. It does not rebound over the next two passages.
The immediate closing edge replicated.
At the turn, later morphological fanout exceeded the same intervention two passages afterward by 0.00290, 95% interval [0.00049, 0.00529]. That is a 13.9% reduction in steerability.
The +2 and +4 states are indistinguishable from one another. The point estimate stays lower at +4; the direct turn-to-+4 interval touches zero by only 0.000017.
Every fixed replication split points the same way.
The peak-to-after loss of leverage is positive in each unseen future background and in each independent schedule. The magnitude varies; the direction does not.
A causal perturbation fans outward through the future.
Immediately after the forced founder event, the two timing conditions are nearly indistinguishable. Their difference accumulates over later passages—the signature expected when the intervention is opening different developmental trajectories, rather than changing one instantaneous measurement.
Spatial form closes while composition stays open.
Morphological steerability
The same founder choice loses substantial leverage over the organism’s later spatial field immediately after the information turn.
Founder-composition contrast
95% interval [-0.0029, +0.0038]. Founder proportions do not undergo the corresponding closure.
Information geometry is tracking the collapse of possible forms.
Our earlier prospective result showed that a prefix-only structural information peak predicted later reorganization. The first causal intervention then suggested that morphology became harder to steer after that turn. This dense map now reproduces the boundary in new stochastic futures and reveals its shape.
Before
Founder choice still opens meaningfully different spatial futures. Development has several reachable forms.
After
The same choice still changes composition, but has less power to redirect later morphology. The developmental basin has narrowed.
Move the turn. Does the boundary follow?
The decisive next experiment is to causally advance or delay the structural information turn, then remap morphological susceptibility. If the commitment edge moves with the information turn rather than staying at the same absolute age, information geometry is not merely reading a developmental clock—it is coupled to the mechanism that closes the future.
Clock hypothesis
Shift the information turn, but commitment still occurs at the old absolute passage.
Mechanistic hypothesis
Shift the information turn, and the loss of causal plasticity shifts with it.
A causal commitment result—not yet a universal law.
This is an exact fixed-panel result across 19 existing organism families, two schedules, and two new stochastic future backgrounds. It replicates across future noise, not across a newly sampled population. It identifies a morphology-specific developmental commitment boundary under founder-directed interventions. It does not by itself prove causal emergence or autonomous replication.