Loading
What changed between revision 1 and revision 3, compared over the article's markdown source with its frontmatter — so a change to status, evidence or reversibility shows up as the edit it is.
Older · revision 1
Import· initial corpus import, not an edit
Initial import of content/whole-brain-emulation.md — the filesystem corpus, unchanged. Not an edit.
17.4 kB
Revision 1
Revision 3
lines 7–13 → 7–136 unchanged lines not shown
categories: ["minds", "foundations"]categories: ["minds", "foundations"]tags: ["emulation", "connectomics", "simulation", "consciousness", "computational neuroscience", "uploading"]tags: ["emulation", "connectomics", "simulation", "consciousness", "computational neuroscience", "uploading"]summary: "A proposed procedure for reproducing a particular mind by scanning one brain's structure in detail, building a computational model from the scan, and running it."summary: "A proposed procedure for reproducing a particular mind by scanning one brain's structure in detail, building a computational model from the scan, and running it."updated: "2026-07-27"updated: "2026-10-29"issues: ["The claim that groups are building dynamical models from the fly connectome carries no source."]issues: ["The claim that groups are building dynamical models from the fly connectome carries no source."]------ lines 29–35 → 29–3515 unchanged lines not shown
scan the structure of one specific brain at sufficient resolution, translate that scan into ascan the structure of one specific brain at sufficient resolution, translate that scan into acomputational model of its components, and run the model fast enough and faithfully enough that itcomputational model of its components, and run the model fast enough and faithfully enough that itproduces the same behaviour the original brain would have produced. It is distinct from buildingproduces the same behaviour the original brain would have produced. It is distinct from buildingan artificial mind from scratch, and distinct from simulating a generic brain — the object is aan artificial mind from scratch, and distinct from simulating a generic brain: the object is acopy of *this* brain, retaining its memories and dispositions. No nervous system of any size hascopy of *this* brain, retaining its memories and dispositions. No nervous system of any size hasbeen emulated in this sense, including the 302-neuron nervous system of the nematode *C. elegans*,been emulated in this sense, including the 302-neuron nervous system of the nematode *C. elegans*,whose wiring diagram has been published since 1986.whose wiring diagram has been published since 1986.lines 39–57 → 39–573 unchanged lines not shown
The idea is usually presented as an engineering problem rather than a scientific one. ItsThe idea is usually presented as an engineering problem rather than a scientific one. Itsproponents argue that no new physics is required, only the extension of three existing capabilities:proponents argue that no new physics is required, only the extension of three existing capabilities:volume microscopy, computational neuroscience, and computing hardware. That framing carries avolume microscopy, computational neuroscience, and computing hardware. That framing carries asubstantive assumption — that there exists some level of biological description below which furthersubstantive assumption: that there exists some level of biological description below which furtherdetail can be replaced by statistical noise without changing the emulated system's behaviour. Thedetail can be replaced by statistical noise without changing the emulated system's behaviour. Theassumption is called *scale separation*, and whether it holds for brains is unresolved.assumption is called *scale separation*, and whether it holds for brains is unresolved. The scale of the object is worth stating plainly. A human brain contains roughly 86 billion neuronsA human brain contains roughly 86 billion neurons and a comparable number of non-neuronal cells,and a comparable number of non-neuronal cells, with synapse counts estimated in the hundreds ofwith synapse counts estimated in the hundreds of trillions.[^azevedo2009] Every one of thosetrillions.[^azevedo2009] Every one of those synapses is a structure whose strength, receptorsynapses is a structure whose strength, receptor composition, and history would have to be capturedcomposition, and history would have to be captured or inferred. No proposal treats this as anythingor inferred. No proposal treats this as anything other than the largest measurement problem everother than the largest measurement problem ever attempted on a single object.attempted on a single object. Whole brain emulation is the technical substrate of [[mind-uploading]]; the two are often conflatedWhole brain emulation is the technical substrate of [[mind-uploading]]; the two are often conflatedbut answer different questions. Emulation asks whether a functionally equivalent process can be runbut answer different questions. Emulation asks whether a functionally equivalent process can be runon other hardware. Uploading asks whether the person goes with it — a question abouton other hardware. Uploading asks whether the person goes with it, a question about[[personal-identity-and-continuity]] that no amount of engineering settles. The philosophical[[personal-identity-and-continuity]] that no amount of engineering settles. The philosophicalpremise both share, that mental states depend on organization rather than material, ispremise both share, that mental states depend on organization rather than material, is[[substrate-independence]].[[substrate-independence]].lines 60–67 → 60–712 unchanged lines not shown
### Scanning### Scanning ```figure{"key": "connectome-tractography", "caption": "What scanning can do today: diffusion-MRI tractography averaged over 20 people, resolving millimetre-scale fibre bundles. Emulation would need the synapse-level wiring of one brain, many orders of magnitude finer."}``` The scan must capture whatever the model needs. At minimum that means the complete synaptic wiringThe scan must capture whatever the model needs. At minimum that means the complete synaptic wiringof the brain — the connectome — including which neuron contacts which, where, and through what kindof the brain (the connectome), including which neuron contacts which, where, and through what kindof synapse. Electron microscopy at nanometre resolution can resolve this, but only destructively andof synapse. Electron microscopy at nanometre resolution can resolve this, but only destructively andonly on tissue that has been chemically fixed, stained with heavy metals, embedded in resin, and cutonly on tissue that has been chemically fixed, stained with heavy metals, embedded in resin, and cutinto sections tens of nanometres thick. Every published mammalian connectome fragment has beeninto sections tens of nanometres thick. Every published mammalian connectome fragment has beenlines 73–79 → 77–835 unchanged lines not shown
A wiring diagram alone specifies almost nothing about dynamics. The translation step must assign toA wiring diagram alone specifies almost nothing about dynamics. The translation step must assign toeach element a model and a parameter set: membrane properties, channel densities, synaptic strengthseach element a model and a parameter set: membrane properties, channel densities, synaptic strengthsand their short-term dynamics, neuromodulatory sensitivities. Some of these can in principle beand their short-term dynamics, neuromodulatory sensitivities. Some of these can in principle beinferred from structure — synapse size correlates with strength — but many cannot, and currentinferred from structure (synapse size correlates with strength), but many cannot, and currentelectron microscopy does not read out molecular composition. Correlative approaches that image theelectron microscopy does not read out molecular composition. Correlative approaches that image thesame tissue functionally before slicing it, as the MICrONS mouse cortex project did, are thesame tissue functionally before slicing it, as the MICrONS mouse cortex project did, are theexisting partial answer, but they work only on a living animal, not on a preserved brain.existing partial answer, but they work only on a living animal, not on a preserved brain.lines 92–106 → 96–11112 unchanged lines not shown
connectivity between brain regions at the top, through analogue population models, spiking networks,connectivity between brain regions at the top, through analogue population models, spiking networks,detailed compartmental electrophysiology, and metabolic and proteomic detail, down to stochasticdetailed compartmental electrophysiology, and metabolic and proteomic detail, down to stochasticmolecular dynamics and, at the bottom, quantum-level simulation.[^sb2008] Each rung down multipliesmolecular dynamics and, at the bottom, quantum-level simulation.[^sb2008] Each rung down multipliesthe storage and computation required by orders of magnitude. The roadmap's central point is not anythe storage and computation required by orders of magnitude. The roadmap's central point concernsparticular estimate but the structure of the bet: emulation is feasible if and only if the necessarythe structure of the bet rather than any particular estimate: emulation is feasible if and only iflevel is high on the ladder. Most computational neuroscientists who engage with the proposal thinkthe necessary level is high on the ladder. Most computational neuroscientists who engage with thethe spiking-network level is too coarse; almost nobody thinks the quantum level is required.proposal think the spiking-network level is too coarse; almost nobody thinks the quantum level isrequired. > [!key] Why the level matters> [!key] Why the level matters> The difference between the coarsest and finest proposed levels is not a factor of ten. It is the> The coarsest and finest proposed levels differ by much more than a factor of ten. The gap> difference between a problem that large computing facilities might approach this century and one> separates a problem that large computing facilities might approach this century from one that> that exceeds any physically plausible machine. The roadmap does not resolve which applies.> exceeds any physically plausible machine. The roadmap does not resolve which applies. ## Origins## Origins lines 120–133 → 125–13813 unchanged lines not shown
{ "year": "1988", "title": "Mind Children", "text": "Hans Moravec sets out both the destructive-scan and the gradual neuron-by-neuron replacement routes, framing emulation as a foreseeable consequence of computing trends." }, { "year": "1988", "title": "Mind Children", "text": "Hans Moravec sets out both the destructive-scan and the gradual neuron-by-neuron replacement routes, framing emulation as a foreseeable consequence of computing trends." }, { "year": "2005", "title": "Blue Brain begins", "text": "Henry Markram launches a project at EPFL to reconstruct cortical microcircuitry in biophysical detail, the first sustained attempt at bottom-up brain simulation at scale." }, { "year": "2005", "title": "Blue Brain begins", "text": "Henry Markram launches a project at EPFL to reconstruct cortical microcircuitry in biophysical detail, the first sustained attempt at bottom-up brain simulation at scale." }, { "year": "2008", "title": "The roadmap", "text": "Anders Sandberg and Nick Bostrom publish a technical report at the Future of Humanity Institute defining the scan-translate-run framework and its levels of detail." }, { "year": "2008", "title": "The roadmap", "text": "Anders Sandberg and Nick Bostrom publish a technical report at the Future of Humanity Institute defining the scan-translate-run framework and its levels of detail." }, { "year": "2013–2023", "title": "The Human Brain Project", "text": "An EU flagship funded at around a billion euros; after an open letter of protest from hundreds of neuroscientists in 2014 it was restructured away from whole-brain simulation and ended having delivered research infrastructure instead." }, { "year": "2013-2023", "title": "The Human Brain Project", "text": "An EU flagship funded at around a billion euros; after an open letter of protest from hundreds of neuroscientists in 2014 it was restructured away from whole-brain simulation and ended having delivered research infrastructure instead." }, { "year": "2024", "title": "A whole adult brain mapped", "text": "The FlyWire consortium publishes the complete synaptic connectome of an adult fruit fly, roughly 140,000 neurons — the first whole brain of an animal with complex behaviour to be reconstructed." } { "year": "2024", "title": "A whole adult brain mapped", "text": "The FlyWire consortium publishes the complete synaptic connectome of an adult fruit fly, roughly 140,000 neurons, the first whole brain of an animal with complex behaviour to be reconstructed." }]]`````` ## What has actually been built## What has actually been built Nothing that qualifies. The largest published biophysical reconstructions are of corticalNothing built so far qualifies. The largest published biophysical reconstructions are of corticalmicrocircuits, not brains: the Blue Brain Project's 2015 model of a rat somatosensory columnmicrocircuits, not brains: the Blue Brain Project's 2015 model of a rat somatosensory columncomprised on the order of thirty thousand neurons and tens of millions of synapses, built fromcomprised on the order of thirty thousand neurons and tens of millions of synapses, built fromstatistical rules rather than from a scan of one animal.[^markram2015] It reproduced somestatistical rules rather than from a scan of one animal.[^markram2015] It reproduced somelines 140–146 → 145–1516 unchanged lines not shown
corresponds to a mind.corresponds to a mind. The nearest thing to a substrate for a real emulation arrived in 2024, when the complete synapticThe nearest thing to a substrate for a real emulation arrived in 2024, when the complete synapticwiring of an adult fruit fly was published — roughly 140,000 neurons with cell-type annotation andwiring of an adult fruit fly was published: roughly 140,000 neurons with cell-type annotation andpredicted neurotransmitter identity.[^dorkenwald2024] Groups have since begun buildingpredicted neurotransmitter identity.[^dorkenwald2024] Groups have since begun buildingwhole-brain-scale dynamical models from it. These are properly called simulations rather thanwhole-brain-scale dynamical models from it. These are properly called simulations rather thanemulations: they use the anatomy of one fly with parameters fitted or assumed, and their outputs areemulations: they use the anatomy of one fly with parameters fitted or assumed, and their outputs arelines 171–178 → 176–18324 unchanged lines not shown
much is being abstracted away at the spiking level.much is being abstracted away at the spiking level. **Chemistry is not wiring.** Neuromodulators diffuse and act on receptors at a distance from any**Chemistry is not wiring.** Neuromodulators diffuse and act on receptors at a distance from anysynapse. Eve Marder's work on the crustacean stomatogastric ganglion — a circuit of about thirtysynapse. Eve Marder's work on the crustacean stomatogastric ganglion, a circuit of about thirtyneurons whose anatomy has been fully known for decades — shows that the same wiring producesneurons whose anatomy has been fully known for decades, shows that the same wiring producesqualitatively different rhythms under different neuromodulatory conditions, and that widelyqualitatively different rhythms under different neuromodulatory conditions, and that widelydifferent parameter sets produce identical output.[^marder2012] A structural scan cannot distinguishdifferent parameter sets produce identical output.[^marder2012] A structural scan cannot distinguishthose parameter sets. Glia, which outnumber neurons in some regions and modulate synapticthose parameter sets. Glia, which outnumber neurons in some regions and modulate synapticlines 196–203 → 201–20817 unchanged lines not shown
> Estimates from within the emulation community typically fall in the second half of this century.> Estimates from within the emulation community typically fall in the second half of this century.> Kenneth Miller, a computational neuroscientist at Columbia, argued in 2015 that even mapping the> Kenneth Miller, a computational neuroscientist at Columbia, argued in 2015 that even mapping the> relevant structure of a human brain plausibly requires centuries at any realistic rate of> relevant structure of a human brain plausibly requires centuries at any realistic rate of> improvement.[^miller2015] The disagreement is not about the physics; it is about how many> improvement.[^miller2015] The disagreement concerns how many unknown parameters the translation> unknown parameters the translation step hides.> step hides, not the physics. ## Relationship to other technologies## Relationship to other technologies lines 211–221 → 216–2267 unchanged lines not shown
containing billions of synapses, and record activity rather than structure; the gap is not one thatcontaining billions of synapses, and record activity rather than structure; the gap is not one thatincremental electrode counts close, and the methods of [[neural-decoding]] recover motor plans andincremental electrode counts close, and the methods of [[neural-decoding]] recover motor plans andattempted speech rather than the standing structure that an emulation would need. Person-specificattempted speech rather than the standing structure that an emulation would need. Person-specificcomputational models of the body, described under [[human-digital-twins]], share the goal of modellingcomputational models of the body, described under [[human-digital-twins]], share the goal ofone individual rather than a generic one, at a resolution many orders of magnitude coarser. Emulationmodelling one individual rather than a generic one, at a resolution many orders of magnitudeis also frequently invoked as a route tocoarser. Emulation is also frequently invoked as a route to [[artificial-general-intelligence]] that[[artificial-general-intelligence]] that sidesteps the need to understand intelligence, and as thesidesteps the need to understand intelligence, and as the mechanism behind speculative economies ofmechanism behind speculative economies of copied minds.copied minds. ## Outlook## Outlook removed, struck through added, underlinedLine numbers count the serialised markdown of each revision, frontmatter included.
227 lines are the same in both revisions, and 147 of them are not shown. Neither revision here is necessarily the current article — read Whole brain emulation for the text this wiki stands behind today.