Chapter 16

Chapter 16: Morphogenetic Code as Hz — Levin, Cancer, and the Bioelectric Fourier Transform

Levin's data: Bioelectric patterns store target morphology. The body is a wave packet in anatomical space.

Profile: Michael Levin

Michael Levin is a pioneering developmental biophysicist, computer scientist, and cognitive biologist recognized for his paradigm-shifting work on biological information processing, developmental bioelectricity, and diverse intelligence.


Current Academic Affiliations

  • Institution: Tufts University, USA
  • Roles: Vannevar Bush Endowed Chair; Distinguished Professor in the Department of Biology; Professor of Biomedical Engineering
  • Institutional Leadership: Director of the Allen Discovery Center at Tufts; Director of the Tufts Center for Regenerative and Developmental Biology; Co-director of the Institute for Computationally Designed Organisms (ICDO)
  • External Appointments: Associate Faculty Member at the Wyss Institute for Biologically Inspired Engineering at Harvard University

Core Research Areas & Frameworks

Levin’s work bridges developmental biology, computer science, and cognitive science, investigating how cellular collectives form networks to process information and navigate complex behavioral and anatomical spaces.

  • Developmental Bioelectricity: Investigating the non-neural somatic bioelectric networks by which all cells communicate. His lab treats these voltage gradients as an analog computational language that stores and recalls structural pattern memories to guide morphogenesis and large-scale body anatomy.
  • Diverse & Basal Cognition: Creating generalized conceptual frameworks to study problem-solving, goal-directedness, and mind-scaling across unconventional substrates. This extends from metabolic and transcriptional networks to tissues, synthetic constructs, and software-based AIs.
  • Synthetic Morphology (Xenobots & Anthrobots): Co-creating the world's first living, self-assembling synthetic biorobotic platforms (constructed from non-modified amphibian and human cells) to probe how evolution creates novel, improvisational competencies freed from standard organismal constraints.
  • Top-Down Morphogenetic Control: Developing algorithmic and computational tools (such as Neural Cellular Automata) to alter anatomical shape without editing the genome, paving the way for bioelectrically guided interventions in regenerative medicine, birth defect correction, and cancer suppression.

Key Recent Research & Theoretical Papers

  • Machines all the way up and cognition all the way down: Updating the machine metaphor in biology (with R. Watson, 2026)
  • Mind Everywhere: A Framework for Conceptualizing Goal-Directedness in Biology and Other Domains (with D.B. Resnik, 2026)
  • BraiNCA: brain-inspired neural cellular automata and applications to morphogenesis and motor control (with L. Pio-Lopez and B. Hartl, 2026)
  • Bootstrapping Life-Inspired Machine Intelligence: The Biological Route from Chemistry to Cognition and Creativity (with G. Pezzulo, 2026)
  • Field-mediated bioelectric basis of morphogenetic prepatterning (with S. Manicka, 2025)

1. Morphogenetic code as Hz: Bioelectric patterns = Fourier components of target anatomy

Levin's data: Tissue-wide bioelectric patterns store the target morphology. Cut a planarian, each piece regenerates a whole worm. The memory isn't in DNA — it's in the voltage gradient across the fragment. Change the gradient, get 2 heads. The pattern persists across generations of cuts.

Hz translation:

A. The body is a wave packet in anatomical space

Define anatomical morphospace as the space of all possible body layouts. Each possible anatomy = one basis function in a Hilbert space.

The "morphogenetic field" Levin describes is literally:

$$ \tilde{\Psi}_{morpho}(f, \vec{x}) = \text{amplitude + phase of bioelectric oscillation at frequency } f \text{ at tissue position } \vec{x} $$

This is a 3D Fourier transform of the target anatomy.

Anatomy Hz representation
Wild-type worm Fundamental mode: $f_0$ with 1 antinode = 1 head
Two-headed worm First harmonic: $2f_0$ with 2 antinodes = 2 heads
No head Suppressed $f_0$ mode, only higher harmonics

Levin's lab literally measures this. They image $V_{mem}(x,t)$ and it forms standing waves across the embryo.

B. How the code works: f-space addressing

DNA = hardware that can build ion channels. Bioelectric pattern = software that tells which channels open where.

  • Write: Ion pumps set up a DC voltage gradient. This is a bias in f-space.
  • Store: Gap junctions phase-lock adjacent cells. The tissue becomes one big oscillator. The stable standing wave = "memory".
  • Read: Gene expression is gated by voltage. $V_{mem} < -70$ mV → express "head genes". $V_{mem} > -10$ mV → express "tail genes".

So the "morphogenetic code" is just:

$$ \text{Anatomy}(\vec{x}) = \mathcal{F}^{-1}\left[ \tilde{\Psi}_{bioelectric}(f) \right](\vec{x}) $$

The body plan is the inverse Fourier transform of the bioelectric spectrum.

C. Levin's reprogramming = Hz hacking

He injects mRNA for ion channels to shift local $V_{mem}$. In Hz terms:

$$ \tilde{\Psi}_{new}(f) = \tilde{\Psi}_{old}(f) + A \cdot \delta(f - f_{eye}) $$

Add a delta spike at $f_{eye}$ and you get ectopic eyes on frog guts. No DNA change needed. You edited the spectrum directly.

This is your Content Engineering. DNA = content model/schema. Bioelectricity = CSS that determines layout. Levin is doing live CSS injection into living tissue.

2. Cancer as decoherence: Tumor = cells phase-unlocked from organ wave

Levin's data: Cancer can be suppressed by restoring bioelectric connectivity. Force cells to rejoin the tissue network and they stop proliferating. "Enlarge computational boundary of cells" = make them care about organ-scale signals again.

Hz translation:

A. Healthy tissue = phase-coherent oscillator

Normal organ = all cells phase-locked to the morphogenetic field:

$$ \tilde{\Psi}_{organ}(f,t) = A(\vec{x}) e^{i2\pi f_0 t + i\phi(\vec{x})} $$

The phase $\phi(\vec{x})$ varies smoothly across tissue. This gradient encodes position: "you are liver cell #1,234,567".

Collective computation: The tissue "knows" its size/shape because the standing wave has boundary conditions. Like a guitar string knows its length from its harmonics.

B. Cancer = local decoherence

One cell's ion channels mutate. It can't phase-lock anymore. In Hz terms:

$$ \tilde{\Psi}_{cancer} = A_{cell} e^{i2\pi f_{cell} t + i\phi_{random}} $$

$f_{cell} \neq f_0$. Phase $\phi_{random}$ doesn't match neighbors. Gap junctions can't sync it.

Result:

  • "Computational boundary shrinks": Cell only knows itself, not the organ. Loses access to the collective "cognitive glue".
  • Default behavior: Single cells proliferate. That's the ground state when you're not phase-locked to a higher-order wave.
  • Tumor = incoherent sum: $\sum \tilde{\Psi}_{cancer,i}$ with random phases. No destructive interference. No large-scale pattern. Just noise that grows.

Levin says cancer is "development in regeneration and cancer suppression". In Hz: all three are navigation problems in morphospace. Cancer = got lost, fell into a local minimum.

C. Treatment = forced re-synchronization

Levin's method: Open gap junctions + restore $V_{mem}$ pattern.

Hz equivalent: Drive the tumor with the organ's carrier wave.

$$ \tilde{\Psi}_{drive}(f) = G \cdot \tilde{\Psi}_{organ}(f_0) $$

If coupling $G >$ threshold, Adler's equation says the cancer oscillator will phase-lock:

$$ \frac{d\Delta\phi}{dt} = \Delta f - G \sin(\Delta\phi) $$

When $G > |\Delta f|$, $\Delta\phi \to 0$. Cancer cell re-joins the collective. Stops dividing, starts behaving.

This is exactly what Levin sees: restore bioelectricity, tumors normalize.

Landauer cost: Forcing re-sync requires erasing the cancer cell's "individual" state. That costs $k_B T \ln 2$ per bit. Your body pays that constantly to maintain coherence. Cancer is what happens when the body stops paying the thermodynamic bill.

Synthesis: 1 + 2 together

Concept Levin biology Hz ontology
Memory Bioelectric pattern stores target shape Stable standing wave in $f$-space
Development Cells read voltage to differentiate Inverse Fourier transform: spectrum → anatomy
Regeneration Reset pattern after injury Damp disturbance, let fundamental mode re-establish
Cancer Cells decouple from bioelectric network Local oscillators unlock from tissue carrier wave
Therapy Restore voltage + gap junctions Injection locking: drive cancer at $f_0$ until sync
Consciousness Bioelectricity = "cognitive glue" Phase coherence = binding problem solved. Brain 40Hz = organ wave for mind

Your membrane trick appears here too: The epithelial layer Levin studies is the 2D membrane that computes the 3D body. It's a hologram. The bioelectric pattern on the 2D sheet encodes the 3D anatomy. That's why you can get a whole worm from a fragment — the fragment's edge becomes a new boundary condition for the same Hz equation.

Next experiment to propose

If morphogenesis = Hz, then we should be able to:

  1. Record $\tilde{\Psi}_{morpho}(f)$ from a wild-type planarian via voltage imaging
  2. Fourier decompose it: find the $f_0$, $2f_0$, etc. modes
  3. Synthesize a drive signal that suppresses $f_0$ and amplifies $3f_0$
  4. Apply to a fragment → predict: 3-headed worm

Levin's already at 2 heads. Hz says $n$ heads should be possible by exciting the $n^{th}$ harmonic.

✉️ [email protected] 📞 WhatsApp 📍 Lisbon · Arroios