Microcosmos was validated across four experiments spanning hand-designed control to fully emergent behavior, exercising both gradient descent and neuroevolution on one substrate

The paper behind Microcosmos (arXiv:2607.02954) exercises its engine (claim-microcosmos-elastic-filaments-differentiable-gpu-fluid) across four experiments arranged along a spectrum from fully specified control to fully emergent behavior:

1. Hand-designed locomotion across Reynolds numbers from ~10³ down to ~1, which matched Purcell's scallop theorem — a physics sanity check that the fluid coupling behaves correctly at low Reynolds number.
2. Gradient-based folding of a 1000-node filament into MNIST digit shapes, a demonstration that the simulation is differentiable end to end (the same property that lets INR models be optimized by gradient descent — claim-siren-2020-implicit-neural-representations-signals-as-functions).
3. NEAT / CPPN neuroevolution, which produced sinusoidal swimming and emergent chemotaxis — evolutionary rather than gradient search on the identical substrate.
4. MAP-Elites quality-diversity search, which discovered a range of gaits along bending/stretching behavioral axes.

The intellectually load-bearing point is that experiments 2 and 3–4 run on the same differentiable engine: gradient descent and open-ended evolutionary search (NEAT, MAP-Elites) share one physics substrate. This is the vault's first note touching the Kenneth Stanley / Sakana AI open-endedness tooling lineage; the existing neuroevolution-adjacent material is backpropagation and automatic-differentiation history (backpropagation-gap), not evolutionary or quality-diversity search, so this note is a seed for a new cluster.

The four results are recorded from the capture's paraphrase of §experiments; no verbatim quote was preserved, so the note is held seedling under [unverified-quote] and the primary re-read is routed to question-verify-microcosmos-benchmarks-and-experiments.