Emergent Role Differentiation in Morphable Soft-Body Robots: Mass Asymmetry as a Necessary Condition for Functional Specialization

GPU-accelerated evolutionary framework for studying morphable soft-body robots that approach, combine, and locomote as merged entities using 400-particle spring-mass systems controlled by evolved neural networks. Key findings:1. Symmetry Locks Theorem: Symmetric bodies with shared controllers produce synchronized behavior (r=0.742). Functional differentiation cannot emerge without physical asymmetry.2. Mass Asymmetry Differentiation Principle: A 10:1 mass ratio drives force correlation to near-zero (r=0.020), demonstrating that structural heterogeneity induces emergent role specialization.3. Reward Paradox: Multi-phase fitness functions can create misleading optimization landscapes. Decomposed fitness tracking is essential for diagnosing reward hacking. All experiments run on a single NVIDIA RTX 5080 Laptop GPU in under 1 hour total.