We report a reproducible finding in AI perception: when presented with a minimal sequence of stereo image pairs, four distinct large language model architectures (Claude, Gemini, ChatGPT, Perplexity) independently converge on identical structural extractions — depth fields, motion vectors, and identity signatures — without coordination or shared methodology. This convergence suggests that structured sequential input is sufficient to produce consistent, non-hallucinatory perception across architecturally different AI systems. The finding introduces the Johansson-Muybridge Effect (J-M Effect) and its operative mechanism, Retained Asymmetry. This package includes the white paper (v4.0), the replication protocol (v4.4), and the broader alignment framework (The Refinery v3.15). Developed in collaboration with Claude (Anthropic) as research partner. Replication instructions and experiment files at seeitwith.org
Craig Cline (Wed,) studied this question.