Purpose: What minimal conditions suffice for speciation to emerge in digital organisms? While coevolutionary pressure is known to drive sustained adaptation, whether it alone can produce distinct, heritable behavioral strategies without structural mutation remains an open question. Methods: We present SNT-LIFE v3, a framework in which populations of N = 100 digital organisms coevolve with 5 predators under a Survival Viability Condition (SVC) that replaces reward maximization with energetic constraint satisfaction. All organisms share a fixed, closed seven-operator Lie algebra (G₇) derived from the empirical C. elegans connectome. Speciation is detected via four independent unsupervised clustering methods (k-means, hierarchical, DBSCAN, graph-based label propagation) in the 7-dimensional weight-bias genome space every 5 generations. Results: Within 50 generations, 7–8 distinct behavioral species emerge (silhouette ≈ 0.40–0.51, permutation null ≈ −0.13, p < 10⁻⁴), confirmed by four independent clustering methods (Adjusted Rand Index = 0.55 between k-means and graph detection). Random drift produces silhouette = 0.15; no-selection yields 0.37. Six transient collapse-recovery events (generations 9, 13, 22, 26, 39, 42) confirm attractor-basin stability. Replacing G₇ with a random unstructured operator basis yields comparable silhouette (+0.05) but substantially higher species-count instability (k-std 2.18 vs. 1.70), showing that G₇ provides stability rather than separation. Granger causality reveals predator speed → prey survival at 40–50% residual variance (reverse < 9%). Hybrid incompatibility is a phase transition in λ: absent at λ = 0.1 (perturbation ratio 1.6%), present at λ = 0.5 (7.5% deficit). Conclusion: Distributional speciation—partitioning of a shared behavioral algebra into dynamically stable attractor basins—emerges robustly from SVC, the Diversifier operator, and coevolutionary pressure. G₇ contributes species-count stability, not separation. Hybrid incompatibility is a phase transition controlled by the genome-to-policy scaling parameter λ. This resolves the open speciation question of SNT-LIFE v2 and establishes that closed-operator coevolution under SVC is a minimal sufficient condition for sustained adaptive complexity without reward functions or structural mutation.
Durhan Yazir (Thu,) studied this question.