In a real sense, yes. Biologist Michael Levin’s lab has shown that cells coordinate using bioelectric signals that store instructions for body shape, and that these patterns can behave like a rewritable memory. This is frontier biology about how bodies build themselves. It is wondrous, peer-reviewed, and genuinely strange. It is also not evidence that thoughts reshape reality.

Key takeaways

  • Cells talk to each other with electricity. Levin’s lab argues these voltage patterns form an instructive layer above the genes, helping decide what shape a body builds (Levin, Pezzulo & LaPalme, 2021).
  • A brief change to a flatworm’s bioelectric state can make it regrow two heads, and that altered body plan stays stored as a hidden, rewritable pattern (Durant et al., 2017).
  • Trained flatworms show signs of retaining what they learned even after losing and regrowing their entire head (Shomrat & Levin, 2013).
  • Xenobots, built from frog stem cells, self-organize into moving “living machines” once their cells are freed from their usual context (Kriegman et al., 2020).
  • The honest line: this work makes the mind-body boundary look blurrier and more interesting than we thought. It still says nothing about manifesting outcomes by intention.

Every so often a piece of science arrives that feels like it should change everything, and the hard part is figuring out exactly what it changes. Michael Levin’s research is one of those. Clips of his work travel widely: flatworms that regrow the wrong anatomy, tiny machines made of living cells, the claim that memory might live below the brain. The wonder is real. So is the temptation to stretch it into a proof of whatever you already hoped was true.

This piece walks the line carefully, because the line is the most interesting part. Levin, a biologist at Tufts University, studies how groups of cells decide what to build, and his answer keeps pointing past the genome toward something electrical and pattern-like. We will look at what the experiments show, hold the awe they deserve, and stay just as clear about where the awe stops. The short version: the science is astonishing, the framing is a frontier still being argued over, and the leap to “so your thoughts heal your body” is one the research declines to make.

Who is Michael Levin, and why are biologists paying attention?

Confidence: the bioelectric signaling layer is established; the “cells as problem-solvers” interpretation is Levin’s proposal, actively debated.

Michael Levin runs a lab at Tufts University studying how cells coordinate to build and repair bodies. His central claim is that genetics tells only part of the story. Above the genes sits a layer of bioelectric signaling, patterns of voltage across cells, that helps decide what anatomy a group of cells will construct.

For most of a century, the working picture has been tidy: genes are the blueprint, proteins are the bricks, and the body assembles itself bottom-up from molecular instructions. Levin’s work complicates that picture while keeping it. According to Levin, Pezzulo, and LaPalme’s 2021 review in Philosophical Transactions of the Royal Society B, cells use differences in electrical potential across their membranes as a kind of shared signaling system, and these patterns carry information about large-scale shape that a single gene leaves unspecified. The genome builds the hardware. The bioelectric layer, in his framing, helps run the software of form.

The reason careful biologists pay attention is that the experiments behind this claim are concrete and reproducible. Bioelectricity is measurable. You can read the voltage, change it, and watch the anatomy respond. What is debated is the bigger interpretation Levin layers on top, the language of memory, decision, and goal-seeking applied to tissue. Keep those two things separate as we go: the data, which is solid, and the framing, which is a live and fascinating argument.

Can a flatworm “remember” its body shape after losing its head?

Confidence: the bioelectric and regeneration results are empirically solid and within-field replicated; the word “memory” is the authors’ framing, not a settled claim about recollection.

In a striking and literal sense, it can. Planarian flatworms are champion regenerators. Cut one in pieces and each piece rebuilds a whole worm. Levin’s lab found that the instructions for which shape to rebuild are held partly in the worm’s bioelectric state, and that this stored target can be rewritten and then inherited through future regenerations.

The key experiment is hard to shrug off. According to Durant and colleagues’ 2017 study in Biophysical Journal, a brief, temporary disruption of a flatworm’s natural bioelectric signaling produced a stable fraction of worms that regenerated with two heads instead of one. The remarkable part came next. When the researchers took the worms that had regrown looking perfectly normal and cut them again, in plain water with no further treatment, a portion regrew two heads once more. The altered body plan had been stored as a hidden pattern, invisible in the worm’s genes, histology, and gene expression, yet ready to resurface. As the authors put it, bioelectric properties “can stably override genome-default target morphology,” and reversing the bioelectric state reset the worms to normal. The pattern was, in effect, a rewritable setting.

Then there is the headline that sounds like science fiction. According to Shomrat and Levin’s 2013 study in the Journal of Experimental Biology, planarians trained to feel comfortable in a textured environment retained signs of that familiarity for at least 14 days, long enough to span a full decapitation and brain regeneration. Worms that had lost their heads and grown new ones showed, the authors carefully wrote, “evidence of memory retrieval in a savings paradigm” after regenerating. The hedge matters: a savings paradigm means a regrown worm relearned faster, not that it consciously recalled a fact. Something about the prior training had outlasted the brain that formed it. Whether to call that “memory” is exactly the interpretive question Levin’s critics and allies are still working through.

What are Xenobots, and are they “designed life”?

Confidence: replicated and extended; use the authors’ hedged “reconfigurable organisms,” not “designed life.”

Xenobots are tiny, self-organizing assemblies built from frog stem cells, designed with help from an evolutionary AI algorithm and then constructed by hand from living tissue. Freed from their normal place in an embryo, the cells get busy. They cooperate into something that moves, navigates, and behaves as a small, novel organism.

The work is genuinely new, and the researchers chose their words with unusual care. According to Kriegman, Blackiston, Levin, and Bongard’s 2020 study in the Proceedings of the National Academy of Sciences, an AI method first designed candidate forms in simulation, and “transferable designs” were then built using a cell-based construction toolkit. The team calls the results “reconfigurable organisms” and “living machines,” deliberately sidestepping the tidier headline of “designed life.” That restraint is the honest scientist showing through, and it is worth honoring.

What makes Xenobots stir something deeper is the lesson underneath. These same skin and muscle cells, in a frog, would have become a frog. Lifted out of that context, given a different electrical and physical environment, they assembled into a form that has never existed in nature. The cells carried a flexibility no one had asked them to use. That plasticity, the capacity of cells to build coherent structures outside their default script, is the throughline connecting the flatworms and the Xenobots, and it is the part that keeps biologists up at night in the good way.

Do non-brain cells really show “learning”?

Confidence: promising but preliminary; a single 2024 study, awaiting replication. A learning-like molecular response, not conscious memory.

This is the newest and most provocative thread, and it should be held the most lightly. A 2024 study reported that a hallmark of learning long thought unique to brains, the advantage of spaced practice over crammed practice, also shows up in ordinary human cells that have nothing to do with the nervous system.

According to Kukushkin and colleagues’ 2024 study in Nature Communications, non-neural human cell lines, engineered to glow when a memory-related molecular pathway switches on, responded more strongly and more durably to stimulation that was spaced out over time than to the same stimulation delivered all at once. The “spacing effect” is one of the best-known findings in the psychology of memory: study spread across days beats the all-nighter. Here it appeared at the level of single cells in a dish, far from any brain. The researchers describe it as a learning-like molecular response, and that phrasing is doing important work. This is chemistry inside a cell behaving in a pattern that resembles how learning works. It is not a cell that knows anything.

A single recent study deserves real caution, and this is exactly the kind of result that needs independent replication before anyone builds on it. What earns it a place here is direction rather than certainty. It joins the flatworms and the Xenobots in pointing at one surprising idea: capacities we filed under “mind” may rest on machinery that runs much lower in biology than the brain. A wonder worth sitting with, and a claim worth checking for years before calling it true.

Does any of this prove manifestation works?

Confidence: the absence of a manifestation link is clear; nothing in this literature points that way.

No, and being clear about that is what lets the wonder stay honest. Levin’s research is about how bodies assemble, store form, and repair themselves through cellular signaling. It studies flatworms, frog cells, and cell cultures. It does not study human intention, and it offers no mechanism by which a thought rearranges matter or heals tissue on command.

It is easy to see why the leap tempts people. If a cell can hold a “memory,” if a body runs on patterns more than on fixed parts, maybe the mind reaches all the way down and reshapes the flesh by belief. The experiments point elsewhere. The bioelectric signals Levin manipulates are local conversations among cells, prodded with drugs and ions in a lab, operating on timescales and through pathways that connect to molecular chemistry rather than conscious wishing. Treating “the body is more dynamic than we thought” as “thoughts control the body” trades a real and careful discovery for a slogan it cannot carry. This is the same dividing line that runs through the whole evidence-based case for what the neuroscience of manifestation actually says: your mind genuinely shapes your attention, your behavior, and, through expectation, parts of your physiology, which is the lesson of the placebo effect taken seriously. It works through you, not by rebuilding cells on command.

So why does this matter, and where is the wonder?

Confidence: the empirical findings are solid; the grand interpretation is open and worth holding with curiosity, not closure.

It matters because it quietly redraws an old map. For a long time the convenient story put mind on one side and matter on the other, with the brain as the lone bridge. Levin’s work, alongside a wider turn in biology, suggests the bridge is wider and older than that, that information, memory-like storage, and something resembling problem-solving show up in the basic machinery of cells.

Hold the result at the right altitude and it is more astonishing for being true. A flatworm carries its body plan as a rewritable electrical pattern. Skin cells, given a new context, build a creature that never existed. A cell in a dish answers spaced signals the way a student answers spaced study. None of this needs a cosmic story to be amazing. The amazement is that ordinary biology turns out to be this inventive. It sits comfortably beside the other genuinely open frontiers, like the question of whether consciousness is fundamental or the early, provocative finding that the body may heal partly on perceived time. These are places where honest scientists say “we do not know yet,” and mean it as an invitation rather than a defeat.

Here is where we land, and where a small practical note belongs. The frontier is thrilling and unfinished, and we get to watch it unfold while holding it as the open question it still is. Meanwhile, the things that reliably change a human life are less exotic and far better evidenced: where you place your attention, how you steady your nervous system, and the actions you take. Noesis is built on that settled ground rather than the speculative edge, which is the honest place to stand while the edge keeps moving. Wonder and rigor work together here. They are the two halves of taking this seriously.

Frequently asked questions

Do cells have memory? In a limited, physical sense, some do. According to Durant and colleagues’ 2017 study, flatworms store body-shape instructions in bioelectric patterns that survive and rewrite across regeneration. According to Kukushkin and colleagues’ 2024 study, non-neural human cells respond more to spaced than to massed signals, the same “spacing effect” seen in learning. Researchers call these memory-like; they are molecular and electrical, not conscious.

What is bioelectricity in Michael Levin’s work? Bioelectricity is the pattern of voltage differences across cell membranes that cells use to coordinate with one another. Levin’s lab argues these patterns form an instructive layer above the genes, helping decide what body cells build, an electrical guide that the genes alone leave underspecified (Levin, Pezzulo & LaPalme, 2021).

Are Xenobots alive, or are they robots? Kriegman and colleagues (2020) call them “reconfigurable organisms” and “living machines.” They are built from living frog stem cells, designed with help from AI, and they move and self-organize. The researchers deliberately avoid the simpler label “designed life.” Xenobots are best understood as a genuinely new category, part organism, part construction.

Does Michael Levin’s research prove manifestation is real? No. It is frontier biology about how bodies assemble and repair themselves through cellular signaling. It says nothing about thoughts shaping external events, and it offers no mechanism for healing or manifesting by intention. The connection is a source of wonder about mind and matter, not proof of anything.

Is the idea that “cells are listening” accepted science? The experiments are peer-reviewed and, within the field, replicated or extended. The larger framing, that cells have “memory” or a kind of basic cognition, is Levin’s interpretive proposal, and it remains actively debated. Hold the data as solid and the philosophy as open.

Sources

  • Durant, F., Morokuma, J., Fields, C., Williams, K., Adams, D. S., & Levin, M. (2017). Long-term, stochastic editing of regenerative anatomy via targeting endogenous bioelectric gradients. Biophysical Journal, 112(10), 2231–2243. https://doi.org/10.1016/j.bpj.2017.04.011
  • Kriegman, S., Blackiston, D., Levin, M., & Bongard, J. (2020). A scalable pipeline for designing reconfigurable organisms. Proceedings of the National Academy of Sciences, 117(4), 1853–1859. https://doi.org/10.1073/pnas.1910837117
  • Kukushkin, N. V., Carney, R. M., Tabassum, T., & Carew, T. J. (2024). The massed-spaced learning effect in non-neural human cells. Nature Communications, 15(1), 9635. https://doi.org/10.1038/s41467-024-53922-x
  • Levin, M., Pezzulo, G., & LaPalme, J. (2021). Bistability of somatic pattern memories: Stochastic outcomes in bioelectric circuits underlying regeneration. Philosophical Transactions of the Royal Society B, 376(1821), 20190765. https://doi.org/10.1098/rstb.2019.0765
  • Shomrat, T., & Levin, M. (2013). An automated training paradigm reveals long-term memory in planarians and its persistence through head regeneration. Journal of Experimental Biology, 216(20), 3799–3810. https://doi.org/10.1242/jeb.087809