Publications
Grounding Without Convention: Parallel Hebbian Drift in Reward-Free Homeostatic Agents as a Positive Control for Iterated-Learning Theory
In a reward-free, homeostatic agent, independent populations ground their signals to a shared meaning-space yet drift into population-specific forms — meaning converges while signal form diverges — and seven pre-registered attempts to elicit a shared arbitrary convention return null. Positioned as a control for iterated-learning theory (Kirby and colleagues).
Accepted — ALIFE 2026 (Late-Breaking Abstract, non-archival)From Empty Graph to Self-Modification via Emergent Preconditions: Growing Cognitive Agents through Dual-Replicator Dynamics
Integration paper: in simulation, agents born with zero memes develop compositional language (TopSim median 0.73), all-or-none cultural convergence, persistent functional meta-cognition (SMC²), and self-modification rules — without reward signals or pretraining. Three ablations identify Cultural Memory as a facilitating mechanism for meta-cognitive persistence (binary ablation 3/9 vs 9/9; NI not established — CM facilitates, not necessary). Ten survival experiments are consistent with language parasiticity on working memory.
Preprint on ZenodoWhy Four? Working Memory Capacity as Evolutionary Equilibrium in a Biomemetic Architecture
Why can humans hold approximately four items in working memory? Using BMC agents with heritable WM capacity and metabolic brain cost in a delayed-reward foraging environment, we show that population mean converges to k≈4.4 from both directions — Cowan's 4±1 range. A robustness–differential tradeoff constrains planning evolution: making planning universally reliable eliminates the selection gradient.
Preprint on ZenodoIt's Not the Silence — It's the Muzzle: Communication Need as Memetic Replication Pressure
Why does isolation cause suffering when survival needs are met? We model communication need as arising from memetic replication pressure (Rexpr) — the M-layer drive to express. Reception without expression is indistinguishable from complete isolation (p = 0.64, N = 79). Expression even into void provides relief. A human validation protocol (N = 24) is proposed.
Preprint on ZenodoFrom Empty Graph to Emergent Language: Ontogenesis of Cognitive Architecture
Agents born with zero knowledge develop cognition and language through domain-general mechanisms. 85–97.5% Lewis signaling game accuracy without gradient-based communication optimization, validated across three embedding spaces (25–533 concepts). Cultural ratchet, survival neutrality, Language Emergence Threshold. Sensitivity analysis: 90 runs support robustness in simulation across >200% parameter variation.
Preprint on ZenodoA Tension Is All You Need: The Biomemetic Complex as a Unified Computational Theory of Consciousness
Entry-point article presenting BMC — consciousness as tension between genetic drives (G) and cultural memes (M), regulated by immune filtration (I) on substrate (S). Polynomial-time CL metric, 5 limiting-case lemmas, a consistent account for all nine COGITATE results (retrodiction, not prediction; rival silence scored N/A, never as a miss). Scaling analysis from N=100 to N=10,000 nodes. The site’s predictions page now lists six bets against named rivals; the broader catalogue this article refers to is archived as a coverage atlas, not as tests.
Preprint on ZenodoResources
BMC for Everyone: A Plain-Language Guide
No equations, no jargon — just the idea and why it matters. Covers all 19 chapters of the theory with diagrams.
Available nowHow to Cite
If you reference BMC Theory in your work:
Zhuravlev, A. (2026). A Tension Is All You Need: The Biomemetic Complex as a Unified Computational Theory of Consciousness. Zenodo. https://doi.org/10.5281/zenodo.18733164
Zhuravlev, A. (2026). From Empty Graph to Emergent Language: Ontogenesis of Cognitive Architecture. Zenodo. https://doi.org/10.5281/zenodo.19181798
Zhuravlev, A. (2026). It’s Not the Silence — It’s the Muzzle: Communication Need as Memetic Replication Pressure. Zenodo. https://doi.org/10.5281/zenodo.19309824
Zhuravlev, A. (2026). Why Four? Working Memory Capacity as Evolutionary Equilibrium in a Biomemetic Architecture. Zenodo. https://doi.org/10.5281/zenodo.19310012
Zhuravlev, A. (2026). From Empty Graph to Self-Modification via Emergent Preconditions: Growing Cognitive Agents through Dual-Replicator Dynamics. Zenodo. https://doi.org/10.5281/zenodo.19559519