Developmental Bioelectricity: the cognitive glue enabling evolutionary scaling from physiology to mind

What to take away

1. 1. Transient pharmacological modulation of Vmem in planarian flatworms permanently resets the bioelectric circuit's homeostatic setpoint so that all subsequent rounds of regeneration produce two-headed animals without any genetic change, demonstrating that somatic bioelectric state constitutes a rewritable morphogenetic memory.
2. 2. Xenopus tadpoles with no primary eyes but ectopic eyes grafted onto their tails can learn a color-vision task using those ectopic organs, which connect to the spinal cord rather than the optic tectum, showing that functional sensory learning does not require evolutionarily canonical neural wiring.
3. 3. Xenobots—proto-organisms assembled from dissociated Xenopus frog embryo skin cells—spontaneously achieve kinematic self-replication by rearranging loose cells in their medium within 48 hours of first creation, a Von Neumann-style replication strategy not known in any other species.
4. 4. Gap junction blockade in wild-type Girardia dorotocephala flatworms stochastically induces regeneration of head anatomies characteristic of species 100–150 million years of evolutionary distance away, with no genetic alteration required, illustrating that bioelectric network states can access normally latent morphospace attractors.
5. 5. Newt kidney tubules maintain a default cross-sectional diameter of 8–10 cells; when cells are experimentally enlarged, fewer cells compensate to preserve tubule diameter, and at extreme cell sizes a single cell bends around itself to achieve the same anatomical outcome, demonstrating downward causation and top-down goal-directedness in morphogenesis.
6. 6. Repeated limb bud amputation in axolotls leads to habituation of the regenerative response and eventual failure to regrow, directly paralleling behavioral habituation in neural systems and supporting the claim that morphogenetic memory shares mechanistic features with neural learning.
7. 7. The paper introduces the Evolutionary Pivot hypothesis, which predicts that the same ion-channel and gap-junction computational hardware used for anatomical morphospace navigation was exapted for behavioral 3D-space navigation when neurons and muscles evolved, with the primary change being a shift from spatial bioelectric patterning (hours timescale) to temporal spiking patterns (milliseconds timescale).
8. 8. To replicate the bioelectric memory-rewriting paradigm, researchers should use fluorescent voltage-reporter dyes for in vivo Vmem imaging in Xenopus embryos or planarians, combined with pharmacological ion-channel modulators (e.g., H,K-ATPase blockers, gap junction blockers such as octanol, or proton–potassium exchanger inhibitors such as SCH28080), applied transiently before injury, then assay regenerative outcomes across multiple amputation rounds.
9. 9. An open question raised by the framework is how many advanced cognitive-science concepts—place cells, path planning, counterfactual simulation, attentional salience—will prove applicable to morphospace navigation, and whether the full panoply of phenomena studied in behavioral neuroscience has direct morphogenetic analogs that await systematic experimental identification.
10. 10. Severe brain malformations in Xenopus induced by chemical teratogens or Notch pathway mutations can be rescued by pharmacological reinforcement of correct Vmem patterns, suggesting that hardware-level genetic defects can be corrected 'in software' via bioelectric intervention and opening a route to regenerative medicine applications that bypass direct gene editing.

Peer brief — for seminar discussion

Michael Levin's 2023 review in Animal Cognition (26:1865–1891) synthesizes experimental and theoretical work to argue that developmental bioelectricity—ion-channel and gap-junction-mediated Vmem dynamics across all somatic cells—constitutes an evolutionarily ancient computational substrate that enabled collective intelligence in morphogenesis before neurons existed, and that the algorithms supporting anatomical problem-solving are homologous to those supporting behavioral cognition. The paper does not report new primary data but instead serves as a framework piece introducing the Multiscale Competency Architecture concept, which holds that each biological organizational level solves problems in its own problem space, with higher levels deforming the energy landscape for lower levels in a downward-causal fashion. The load-bearing empirical pillars are drawn from Levin's own program and allied work: planarian Vmem perturbation experiments (using H,K-ATPase blockers and gap junction inhibitors such as octanol and SCH28080) that permanently rewrite the number-of-heads homeostatic setpoint across subsequent regeneration rounds without genetic change; Xenopus tadpoles with functional ectopic tail eyes that support color-vision learning despite novel spinal-cord connectivity; Xenobots achieving kinematic self-replication within 48 hours; and newt kidney tubules maintaining normal diameter even when a single enlarged cell wraps around itself to hit the anatomical target. These are unified under the Evolutionary Pivot hypothesis: the same ion-channel and gap-junction machinery that originally navigated anatomical morphospace was exapted, when nerve and muscle appeared, for behavioral navigation of 3D space—differing principally in timescale (spatial standing-wave patterns over hours versus temporal spike patterns over milliseconds). A complementary alternative framework the paper could have engaged more systematically is active inference / free-energy minimization (Friston et al.), which formalizes goal-directedness in hierarchical generative models; Levin cites it but does not derive testable predictions from the comparison. The framework implies that the full conceptual toolkit of behavioral neuroscience—memory, perceptual bistability, false-memory induction, neural decoding, attentional recruitment—maps onto morphogenetic phenomena, making developmental biology and cognitive science two dialects of the same language. Practically, it predicts that bioelectric behavior-shaping of cell collectives will outperform bottom-up genetic micromanagement for regenerative medicine, and that tumorigenesis can be controlled by reconnecting cancer cells to tissue-level bioelectric networks rather than through cytotoxic approaches. A critical reader would push back on the evidentiary asymmetry at the heart of the argument: the morphogenetic 'cognition' claims rest almost entirely on a single model system (planarians, chiefly Levin's own lab) plus a handful of Xenopus and axolotl experiments, while the generalization to a universal Multiscale Competency Architecture spanning bacterial biofilms, Physarum, plant electrophysiology, and synthetic Xenobots is largely conceptual extrapolation. The conceptual mappings in Table 1—equating, for instance, 'forgetting' with cancer or 'addiction' with nerve-dependent regeneration—are evocative but unfalsified; no quantitative criteria are offered for when a morphogenetic behavior does or does not qualify as a genuine cognitive analog. The Evolutionary Pivot hypothesis is plausible but currently lacks the phylogenetic comparative data or functional genomic evidence that would distinguish it from the alternative that neural and non-neural bioelectric systems share components through deep homology without the algorithms themselves having been pivoted. These are tractable empirical questions, and the framework's value lies precisely in making them tractable, but the review presents the architecture as more established than the data presently support.

Frameworks (2)

• Basal Cognition
  An interdisciplinary research framework that reconceptualizes intelligence as observer-relative problem-solving competencies existing on a continuum from simple to highly complex, extending cognition beyond neural systems to pre-neural and non-neural substrates including microbial control loops, plants, tissues, and cellular collectives. It grounds the study of evolutionary and developmental origins of cognitive and behavioral capacities by linking information processing at the chemical and cellular level to classical cognition, and provides philosophical foundations for understanding agency and goal-directedness in systems without nervous systems.
• Multiscale Competency Architecture
  A framework originating from Levin that formalizes how hierarchical biological systems—from cells to tissues to organs—exhibit integrated problem-solving and adaptive plasticity across multiple levels of organization (metabolic, transcriptional, physiological, anatomical). It models system-level behaviors as emergent from competition and cooperation among heterogeneous subunits within composite agents, explaining how goals and regulations scale across biological scales.

Claims (20)

• All intelligence is collective intelligence: each of us consists of a huge number of cells working together to generate a coherent cognitive being with goals, preferences, and memories that belong to the whole and not to its parts.

  Core interpretive thesis of the paper.

• Gap junctions provide an 'owner wiping property' for signaling molecules, leading to partial erasure of individual identity for cells in a network and enabling a higher-order emergent collective.

  Mechanistic claim for how collectives scale.

• Biomedicine has not yet internalized the multi-scale approach; behavior-shaping paradigms for cells and tissues will enable much greater control of morphology than bottom-up micromanagement.

  Practical vision for regenerative medicine.

• The decoupling of the material state (protein content) from the information content is the first step to the most amazing aspect of neural networks: they enable mind to arise from matter.

  Argues for a hardware/software distinction enabling emergence of mind.

• Planarian bioelectric circuits can enter a metastable cryptic state where they stochastically produce one- or two-headed worms upon each amputation, analogous to perceptual bistability.

  Links cognitive and morphogenetic dynamics.

• Newt kidney tubule cells, when artificially enlarged, can bend a single cell around itself to achieve correct tubule diameter, illustrating top-down control over molecular mechanisms.

  Extreme example of regulative flexibility.

• When a subset of cells is given a morphogenetic goal (e.g., build an eye), they autonomously recruit unmodified neighbors to complete the structure, showing modular competency.

  Principle of top-down control.

• The same wild-type genome can produce 1-, 2-, or 0-headed planaria depending on bioelectric history, showing that anatomical outcome is not directly encoded in the genome.

  Demonstrates the role of epigenetic bioelectric software.

• Xenobots perform kinematic self-replication, a novel behavior not known in any other species on Earth, using a Von Neumann-style replication within 48 hours of creation.

  Highlights extraordinary plasticity and problem-solving.

• Pattern memory in planaria can be re-written non-genetically via transient modification of membrane potential, leading to permanent changes in target morphology.

  Summarizes the two-headed planarian phenomenon.

Questions (2)

• Could recognizing Diverse Intelligence (in unfamiliar substrates) be essential to understanding the mechanisms and origins of conventional cognition and behavior?

  Motivates expansion of cognitive science.

• What is the coordination mechanism that enables this cross-level integrated information processing?

  Key mechanistic question answered by developmental bioelectricity.

Related work— refs + corpus + external arXiv

Cited / in-corpus / arXiv badges show which signals surfaced each row. Multi-source rows weighted higher.

• The computational boundary of a 'self': developmental bioelectricity drives multicellularity and scale-free cognition

  cited

  in corpus

  2019
  ≈ 92%
• Technological Approach to Mind Everywhere: An Experimentally-Grounded Framework for Understanding Diverse Bodies and Minds

  cited

  in corpus

  2022
  ≈ 86%
• Darwin's agential materials: evolutionary implications of multiscale competency in developmental biology

  in corpus

  2023
  ≈ 88%
• Endogenous Gradients of Resting Potential Instructively Pattern Embryonic Neural Tissue via Notch Signaling and Regulation of Proliferation

  cited

  2015
  ≈ 87%
• The collective intelligence of evolution and development

  in corpus

  2023
  ≈ 87%
• AI-driven control of bioelectric signalling for real-time topological reorganization of cells

  Gon\c{c}alo Hora de Carvalho

  2025
  ≈ 86%
• Endogenous Bioelectric Signaling Networks: Exploiting Voltage Gradients for Control of Growth and Form

  cited

  2017
  ≈ 86%
• Collective intelligence: A unifying concept for integrating biology across scales and substrates

  in corpus

  2024
  ≈ 85%
• The biogenic approach to cognition

  in corpus

  2005
  ≈ 85%
• Transmembrane voltage potential controls embryonic eye patterning in <i>Xenopus laevis</i>

  cited

  2011
  ≈ 85%
• AI: a Bridge toward Diverse Intelligence and Humanity’s Future

  in corpus

  2024
  ≈ 85%
• The brain is required for normal muscle and nerve patterning during early Xenopus development

  cited

  2017
  ≈ 85%
• Basic principles drive self-organization of brain-like connectivity structure

  Valeriy A. Makarov Slizneva, and Cees van Leeuwen Carlos Calvo Tapia

  2024
  ≈ 84%
• Endless forms most beautiful 2.0: teleonomy and the bioengineering of chimaeric and synthetic organisms

  in corpus

  2023
  ≈ 84%
• Microelectronic Morphogenesis: Progress towards Artificial Organisms

  Daniil Karnaushenko, Minshen Zhu and Oliver G. Schmidt John S. McCaskill

  2024
  ≈ 84%
• Self-Improvising Memory: A Perspective on Memories as Agential, Dynamically Reinterpreting Cognitive Glue

  in corpus

  2024
  ≈ 84%
• Neural Manifolds as Crystallized Embeddings: A Synthesis of the Free Energy Principle, Generalized Synchronization, and Hebbian Plasticity

  Vikas N. O'Reilly-Shah

  2026
  ≈ 84%
• Re-membering the body: applications of computational neuroscience to the top-down control of regeneration of limbs and other complex organs

  cited

  2015
  ≈ 84%
• Modeling Bioelectric State Transitions in Glial Cells: An ASAL-Inspired Computational Approach to Glioblastoma Initiation

  Wiktoria Agata Pawlak

  2025
  ≈ 84%
• Morphological Coordination: A Common Ancestral Function Unifying Neural and Non-Neural Signaling

  cited

  2019
  ≈ 83%
• The Physical Basis of Prediction: World Model Formation in Neural Organoids via an LLM-Generated Curriculum

  Brennen Hill

  2025
  ≈ 83%
• Endogenous Voltage Potentials and the Microenvironment: Bioelectric Signals that Reveal, Induce and Normalize Cancer

  cited

  2014
  ≈ 83%
• Gap junctional signaling in pattern regulation: Physiological network connectivity instructs growth and form

  cited

  2016
  ≈ 83%
• Generalizing frameworks for sentience beyond natural species

  in corpus
  ≈ 83%
• Physiological inputs regulate species-specific anatomy during embryogenesis and regeneration

  cited

  2016
  ≈ 83%
• Metabolic basis of brain-like electrical signalling in bacterial communities

  cited

  2019
  ≈ 83%
• Bioelectrical Interfaces Beyond Cellular Excitability: Cancer, Aging, and Gene Expression Reprogramming

  Matthew Burgess, Catarina Franco Jones, Titouan Luciani, Marzia Iarossi, Manuel Schr\"oter, Nako Nakatsuka, Mustafa B. A. Djamgoz, Gil Gon\c{c}alves, Paola Sanju\'an-Alberte, Paula M. Mendes, Frankie J. Rawson, Malavika Nair, Michael Levin, Rosalia Moreddu Paolo Cadinu

  2025
  ≈ 83%
• Behaviorist approaches to investigating memory and learning: A primer for synthetic biology and bioengineering

  cited

  2021
  ≈ 83%
• Dissociated Neuronal Cultures as Model Systems for Self-Organized Prediction

  Zhuo Zhang, Dai Akita, Tomoyo Isoguchi Shiramatsu, Zenas Chao, Hirokazu Takahashi Amit Yaron

  2025
  ≈ 83%
• Booting up the organism during development: Pre-behavioral functions of the vertebrate brain in guiding body morphogenesis

  cited

  2018
  ≈ 82%

+27 more

Similar preprints — Semantic Scholar

Cited by (6)

• Stress sharing as cognitive glue for collective intelligences: A computational model of stress as a coordinator for morphogenesis

  Stress sharing between cells accelerates collective morphogenesis and expands individual cells' cognitive light cone, as demonstrated in a multiscale agent-based computational model of embryogenesis p

• Topological constraints on self-organisation in locally interacting systems

  Topology of local interactions is the decisive factor determining whether a system can sustain long-range order, and decoder-only transformer architectures are provably unable to maintain such order f

• Collective intelligence: A unifying concept for integrating biology across scales and substrates

  Collective intelligence, understood as William James' capacity to reach the same goal by different means, operates not only in beehives and ant colonies but as a scale-free organizing principle across

• Topological constraints on self-organization in locally interacting systems

  Topological constraints on interaction graphs determine whether a locally interacting system can sustain long-range ordered phases, and therefore whether it can self-organize toward a system-level goa

• Brains and where else? Mapping theories of consciousness to unconventional embodiments

  Rouleau and Levin's 2026 analysis in *Philosophical Transactions of the Royal Society A* (384, issue 2320) demonstrates that the functional and operational principles underlying most contemporary theo

• Cognitive glues are shared models of relative scarcities: the economics of collective intelligence

  Levin and Lyons argue that the price system in market economies is not merely a resource-allocation device but the canonical instantiation of what they term a *cognitive glue* — a coordinating afforda