How can you protect a brain by destroying it?


Last week, Simon Davis wrote to me with questions about this cryonic brain preservation technique, which has now been published as How to Freeze Your Brain and Live Forever (Maybe). Unfortunately, my comments did not make it into the story, because, Simon politely explained, there are length restrictions and perhaps, I assume, also because my extended dismissive scorn does not translate well to polite journalism. And that’s OK! Because I have a blog, and I can rant here!

The Brain Preservation Society has a goal: to preserve dead brains today, so they can be reanimated at some distant time in the future. At least, that’s what they say — I’m more inclined to believe their goal is to pocket lots of money exploiting people’s fear of death. Their immediate plan, though, is to develop more thorough mechanisms of locking down the fine structure of brains.

Extending today’s existing small-volume neural preservation techniques to whole brains is essential to the scientific goal of mapping neuronal connectivity across an entire human brain – a goal that has been identified by the NIH and others as crucial to furthering of our knowledge of brain function – see for example the NIH’s Human Connectome Project. Furthermore, advances in neuroscience today strongly suggest that appropriately preserved brains will contain our memories, identity, and a substrate for future consciousness, so that an appropriately verified preservation technology may allow future reanimation of the memories and identity of the preserved individual, if desired.

Lovely. I’m all in favor of mapping neuronal connectivity — I did some small-scale stuff in that field decades ago. But they make extravagant claims.

The technology they are using is straightforward and useful.

We describe here a new cryobiological and neurobiological technique, aldehyde-stabilized cryopreservation (ASC), which demonstrates the relevance and utility of advanced cryopreservation science for the neurobiological research community. ASC is a new brain-banking technique designed to facilitate neuroanatomic research such as connectomics research, and has the unique ability to combine stable long term ice-free sample storage with excellent anatomical resolution. To demonstrate the feasibility of ASC, we perfuse-fixed rabbit and pig brains with a glutaraldehyde-based fixative, then slowly perfused increasing concentrations of ethylene glycol over several hours in a manner similar to techniques used for whole organ cryopreservation. Once 65% w/v ethylene glycol was reached, we vitrified brains at -135 °C for indefinite long-term storage. Vitrified brains were rewarmed and the cryoprotectant removed either by perfusion or gradual diffusion from brain slices. We evaluated ASC-processed brains by electron microscopy of multiple regions across the whole brain and by Focused Ion Beam Milling and Scanning Electron Microscopy (FIB-SEM) imaging of selected brain volumes. Preservation was uniformly excellent: processes were easily traceable and synapses were crisp in both species. Aldehyde-stabilized cryopreservation has many advantages over other brain-banking techniques: chemicals are delivered via perfusion, which enables easy scaling to brains of any size; vitrification ensures that the ultrastructure of the brain will not degrade even over very long storage times; and the cryoprotectant can be removed, yielding a perfusable aldehyde-preserved brain which is suitable for a wide variety of brain assays.

But first of all, you need to understand that their current methods don’t involve simply freezing brains. They are freezing brains and perfusing them with glutaraldehyde to fix them, better preserving the ultrastructure of the tissue. Every microscopists does this; I was fixing zebrafish brains with a cocktail of acrolein, glutaraldehyde, and paraformaldehyde in the 1980s, trying to capture detailed images of synapses in the spinal cord. It worked pretty well, too. They know what they are getting out of this: an aldehyde-preserved brain.

What I didn’t do with my experiments in aldehyde-preserved brains was claim that I was preserving all the information necessary for nervous system function. I was quite aware that I was chemically nuking all the proteins in the tissue; I was washing out most of the chemistry; I was destroying most of the physiological information to preserve a structural skeleton of what was there, so I could see the physical arrangement of the pieces. Nothing more.

Neuroscience does not suggest that fixing a brain in aldehydes will preserve “memories, identity, and a substrate for future consciousness”. There’s no reason to think their methods create “appropriately preserved brains”.

There is a respectable question to be answered with their techniques about the structure of the brain, but it is only one tiny step forward in understanding how the brain functions and generates a mind. Ultrastructure is something we can study, so it’s an issue of chasing the question that can be answered while telling everyone you’re trying to address a completely different question that can’t.

The pattern of synaptic connections is an essential part of the story, but it is not sufficient. A simple counterexample: consider the effect of MAO inhibitors and various antidepressants. They modulate the activity of the brain by affecting the intercellular concentration of neurotransmitters. Consider hormonal effects: your brain is profoundly altered by the chemical signals in your blood (and recursively, secretes hormones of its own). There’s this whole phenomenon called non-synaptic plasticity, in which the behavior of ion channels and pumps is modified by, for instance, phosphorylation, or binding of cofactors.

These things are all destroyed by fixation. The information is no longer there.

If this organization is so confident that they have preserved all the necessary information, I’d like to know why they’re playing around with just the first step of the problem, doing so in impractically complex organisms, and not working on the necessary step of recovering that information. That’s the real test.

Take a simpler organism, like a fruit fly or a nematode. Kill it, fix it, freeze it, vitrify it. That should be trivial at their tiny scale. Then rebuild a fly brain from the extracted information and show me that it still knows how to walk, fly, eat, court, and mate.

Nobody’s even close to accomplishing any of that.

Until then, any talk of an adequate preservation method is simply wishful thinking, especially when it relies on the kind of obliteration of the molecular information in the brain that the Brain “Preservation” Society is doing.

Of course, I actually know why they don’t do any of that. Fruit flies and nematodes won’t pay them a substantial annuity to have their brains vitrified and stored, and their gratitude upon being resurrected wouldn’t be at all remunerative.

But for now, they’ve got really good EM technique and can show off pretty pictures of well-fixed cells. Bravo! Who knew that I was so cutting-edge 30 years ago?


  1. Nerd of Redhead, Dances OM Trolls says

    Glutaraldehyde will non-specifically cross link amino groups, and is probably is not easily reversible (it is used for binding enzymes to substrates for continuous flow enzymatic processes). You have destroyed the integrity of the sample with this cross linking.

  2. Bernard Bumner says


    Yes. Even the best CLEAs or immobilised enzymes will: a) have lost significant activity versus free enzyme; b) leach protein over time because of incomplete cross-linking. And you can’t unmake CLEAs.

    The idea that these people are preserving function is laughable.

  3. Nick Gotts says

    But haven’t you watched Young Frankenstein? Harness the revivifying power of a sufficiently spectacular electrical storm, in the correct environment (the repurposed dungeon of a Transylvanian castle, with plenty of flasks of brightly-coloured, fuming liquids, and an assistant with protruding eyes and severe scoliosis), and Bob’s your uncle (again)!

  4. Matt G says

    Yes, brains absolutely need to be frozen – to be fed to the zombies so we can eke out a few more days before our annihilation.

  5. says

    For some reason I am reminded of the Lake Wobegon incident where one of the contributions to the church potluck was an unfortunately realistic Jell-O mold of a human brain, even to the pinkish=gray color. No one partook of the brain “although it was clear a lot of thought went into it.”

    That may not be the case in this instance.

  6. jimb says

    Marcus Ranum @ 13:

    I believe that’s the name of a Rapid Offensive Unit. I’m pretty sure, anyway.

    Well if it’s not, it should have been. :-)

  7. Gregory Greenwood says

    OK; can someone please tell the the Brain Preservation Foundation that Ghost in the Shell is not – repeat; is not – a documentary?

    No amount of chemical fixing of the general structure of a brain is going to let you resurrect a dead person or transfer yourself into an android body some day. It just isn’t going to happen in the foreseeable future, and they need to accept that and move on.

  8. magistramarla says

    I think that it will do more good to donate my brain to dystonia research, since I have Spasmodic Dysphonia (like Diane Rehm on PBS). Hopefully, that research can help future generations to not have the hassles that I deal with on a daily basis.

  9. brett says

    If you do believe in brain emulation happening some day, then rather than trying to preserve your physical brain, wouldn’t it make more sense to have scientists do a bunch of scans on your brain to try and map the synapses in your brain and nervous system? Put that on some type of extremely durable medium along with instructions in several languages, and then find a way to store it with requests that emulation folks down the line try and reconstruct you from it.

    It’s not really immortality – as far as I’m concerned, your consciousness and sense of self is an embodied process, total cessation of that process is death, and making a copy of you to go on afterwards isn’t really a continuation. But it’s something.

  10. Bernard Bumner says


    Is that sufficient information? Presumably, the fact that mind is conscious and dynamic without each and every connection changing constantly speaks about the fact that the electrochemical state of your brain is also critical? On the other hand, we also know that the plasticity of those connections is critical to function, so a snapshot won’t recapitulate your mind over time.

    Those knowledgeable about neurobiology can comment, but my guess is that a map of synapses (even if such a thing can be captured)
    is a very much incomplete picture of a mind.

  11. Rowan vet-tech says

    I just have to wonder… say we *are* able to preserve these brains adequately. Considering current population trends, *why* would future humans ever resurrect past dead ones? There will be plenty of current humans, no need for old relics.

  12. says

    I am not a neuroscientist, but…
    back when I was studying for my undeserved and unused BA in psychology, I learned that it appeared that a lot to do with memory was the potential (i.e.: approximately, the probability of firing across a given connection) between nerves. That has something to do with there existing a connection, but also with its firing frequency and other squishy chemical stuff we didn’t understand back in 1985. Don’t these guys understand that if you’ve got a computer, with all the wires in it, you still have nothing without the BIOS memory? It’s not simply a matter of knowing there’s an axon from A to B it’s a question of “what is the probability that B will fire given A and … through X?” In computer-ese, if I give you a frozen copy of my computer, and you forgot to record that it’s running Diablo III that’s one thing (you forgot the RAM and peripheral state!) but what about the game state? (I was in the Lair of Damned Cussedness) and detailed game state (about to open the Chest of Damned Cussedness) predictive game state (the chest was going to contain the Scepter of Dawkins) and device state (my Kensington trackball was at 3,211X `2733Y and my Sony flat panel is running in 4k mode)

    What irritates me about these technofetishist fantasies is that they are perpetrated by geeks, who ought to understand the problem better even in terms of their own simplifying abstractions. It’s not anywhere as simple as they make it sound, which sets all my scam alerts ringing.

    Even if it was as simple as that, let me translate: “you’re capturing the node-graph of a markov chain without the inter-node probabilities”; we’re done. It’s actually the probabilities that do the work.

  13. says

    *why* would future humans ever resurrect past dead ones?

    That’s a fun one, too. You can resurrect it as a thought experiment:
    If your cellphone one day claimed to be an entity from the past and demanded better access to the internet, more bandwith, and said it was a racists MRA — would you:
    a) reboot it and hope it got better
    b) listen to it, assume it had passed the turing test and
    c) worship it as my master

  14. applehead says

    I’m surprised that Libertechbro furry guy who stunk up the comments over on “Effective Altruism” hasn’t showed up yet to defend this pile of cynical nonsense.

    I don’t even share PZ’s faint hope there might come something scientifically useful out of their organization. When transhumanists/Singularitarians don’t translate antediluvian religious concepts into a package fit for the digital age (migration of the soul to the afterlife => mind-uploading, resurrection of the dead => perfect brain preservation), they’re busy retroactively declaring real-life technologies as transhumanist/Singularitarian achievements such as IVF, even though not a single scientist or technician associated with their invention ever identified as a transhumanist/Singularitarian.

    On that account these people don’t solely prey on those afraid of their own mortality, but also on a sizeable number who are susceptible to marketing that describes product or service X as a well-known SF trope become reality (see nanotech, private space flight and VR).

  15. says

    Assume we did ressurect jesus of nazareth, and compare his maunderings to a carefully curated philosophy undergrad’s how would he fare? How would he fare against Spinoza or Nietsche let alone Rawls? While it’s fun to picture its “Bambi verus Godzilla” really.”

  16. Azkyroth, B*Cos[F(u)]==Y says

    Why, the same way you destroy a village in order to save it, of course!

  17. anthrosciguy says

    Does their procedure come with a money back guarantee?

    Absolutely. Anyone whose procedure fails may apply and their money will be cheerfully refunded.

  18. lsparrish says

    I just see PZ repeating the same mistake he has always made on this topic in different words. He thinks every bit of information that the brain depends on for function is identity-critical and non-redundant. Cryonics would be a non-starter if that were obviously the case, but it clearly isn’t.

    Aldehyde fixed tissues are not going to provide sufficient data to create models of every detail of how living tissue works. You still do need other approaches in order to be able to determine that, and I don’t see anyone claiming to the contrary.

    But once you have obtained that model, you still have a problem for the goal of reviving anyone in particular, which is that the number of possible brains you can make with it is absurdly high, resulting in almost no chance of reconstructing the essential patterns of the same individual that originally existed.

    This second problem is what cryonicists have judged to be tractable given chemically and/or cryogenically preserved brain tissue as an information source.

    Pretending that we could model organisms today if this really worked is just a really stupid thing to say because it blithely ignores the dependency of the proposal on advanced neural models and computational hardware that doesn’t already exist (and is not predicted by anyone to already exist).