The information arrow flips
as the organism appears.
The same intervention can separate the local information score at every age. Before the frozen visible transition, its high-information direction tends to close future possibilities. After the transition, that direction tends to open them.
The score is controllable throughout development. Its meaning is not.
At every age, the selected high−low whole-minus-sum gap is large and positive: +0.81 to +1.64. Yet the independent future effect is negative around q24–q32, positive by q40, and near zero later.
q24, q40, q48, and q56 were unseen when fixed. q32 is the preexisting calibration that told us where to look.
Age was hiding the real boundary.
At q44, pre-transition organisms still point toward narrowing (-0.0036), while organisms that have crossed the visible boundary point strongly toward expansion (+0.0162). The same split remains at q52.
The event classifier was frozen before this analysis. Using it to divide the causal effects this way is the new, hypothesis-generating discovery.
The direction changes across its own transition.
Information constrains
The high branch tends to occupy a smaller cloud of later futures.
Information diversifies
The high branch tends to occupy a larger cloud of later futures.
Across the 11 organisms that crossed during the scan, the paired change is +0.0210, 95% interval [+0.0040, +0.0422]. Seven move positive; four do not. The cohort shift is coherent, not universal.
We may be seeing a causal reorganization of possibility.
Higher integration, fewer futures
The system behaves as if its parts are becoming committed to a shared developmental path.
Higher integration, more futures
Once organized, the same information direction may support flexible expression rather than commitment.
That is much closer to causal emergence than a generic rising score: an information direction whose causal relationship to the space of possible futures changes at the organism’s visible birth.
We have not yet prospectively confirmed the sign flip on new organisms. That is the next direct test—not a reason to flatten the discovery, but the clean way to find out whether it is a reusable law.
ethereal-pattern-1326 crosses near q43.
q36 → q44, spanning the detected event.
High left, low right. Future effect -0.0553.
High left, low right. Future effect +0.0495.
Freeze the sign flip. Test fresh organisms.
On a completely unseen cohort, let the existing morphology detector choose each organism’s last pre-transition and first post-transition states. At both states, select the information-high and information-low actions from the same eight-action family, then independently measure future repertoire.
High minus low future repertoire is negative at the last pre-transition state.
The same contrast shifts positive across the organism’s own transition.