Tuesday, July 13, 2010

Chemical brain preservation: cryonics for uploaders

Republished in part by the Cryonics Institute (PDF)

The Brain Preservation Foundation and the Brain Preservation Technology Prize were announced at the recent H+ Summit. The week before Ken Hayworth gave a fascinating preview at the ASIM2010-1 First Online Workshop on Advancing Substrate-Independent Minds. Hayworth, a brain researcher at Harvard University, has spent years developing techniques to scan the brain and record its synaptic circuits.

The Brain Preservation Foundation has been established to promote serious scientific research in the field of brain preservation for long-term static storage. Its goal is to spur development of a hospital surgical procedure which can reliably and demonstrably preserve the structural connectivity of 99.9% of the synapses within a human brain. Existing scientific literature suggests that such a goal should be readily achievable by extending (via vascular perfusion) existing laboratory protocols for the chemical fixation and plastic-embedding of small pieces of brain tissue. If such a surgical procedure were available in hospitals it could provide interested persons a means of avoiding death and reaching the distant future. One of the first initiative of the Foundation is the Brain Preservation Technology Prize, a prize for demonstrating ultrastructure preservation across an entire large mammalian brain verified by a comprehensive electron microscopic survey procedure.

Brain preservation is a form of cryonics: preserving dead persons until future technology can restore them to life. In this sense, brain preservation is not an alternative to cryonics, but on the contrary it is conceptually equivalent to cryonics. The chemical brain preservation, or plastination, technique favored by the Brain Preservation Foundation is different from the cryopreservation technique cuttently used by the three major operational cryonics organization (Alcor, the Cryonics Institute and Kryorus), but the objectives are the same. Therefore, I think plastination should be actively investigated by cryonics service providers with the objective of including it as a supported cryonic service when the time is right. The prefix cryo- is not applicable because storing chemically preserved brains does not require ultra low temperatures, but I think the terminology should be preserved to show respect, recognition and gratitude to the cryonics community for their excellent work.

According to Hayworth, a chemically preserved, plastic-embedded brain can be losslessly subdivided in strips that can be imaged at nanometer resolution by current technology. This resolution is sufficient to image the smallest brain structures which, according to current scientific knowledge, are the physical substrate of our thoughts, memories, feelings, emotional responses, hopes, dreams and identity. It is important to stress that this can be done with current technology, and Hayworth cites experimental results to prove it.

So, where is the catch? The catch is that chemical preservation is irreversible, or at least very difficult to reverse (some skeptics think that also conventional cryonics is irreversible). But on the other hand, the information in a chemically preserved brain can be retrieved and run on a different substrate ("mind uploading"). This makes chemical brain preservation a storage technique optimized for future nanoscale scanning, and an ideal form of "cryonics for uploaders". For those who accept scanning the brain and running the information in the scan file on a different substrate as a valid form of identity preservation, chemical brain preservation seems clearly superior to cryopreservation.

I definitely belong to this group. I look forward to being extracted from my brain, leave biology behind, and run as substrate-independent software in a virtual body, roaming the universe and perhaps wearing a physical body on occasions. Mind uploading is a two steps process, 1) scanning the brain to read the information in it, and 2) running the information in the scan file (mindfile) on a suitable alternative substrate. Of course, after the first scanning step 1), the resulting mindfile can be stored until a suitable technology is available for 2). But I used to think that even developing suitable brain scanning technology would take many decades and perhaps centuries. On the contrary, I am now persuaded that chemical brain preservation may permit storing "solid mindfiles", physical databases of memory and personal identity, which future technologies will be able to ignite and bring back to life in a suitable physical substrate, in only one or two decades.

So, I will choose chemical brain preservation once it is operationally available.

Hayworth's "uploading may be only 15 years in the future" (in the sense of operational brain preservation suitable for future scanning and uploading) is very refreshing compared to the cautious, boring and defeatist attitude of today's moderate transhumanists, repented ex-transhumanists and anti-transhumanists in disguise. I feel back in the optimistic 90s, and I hope it lasts. I prefer not to discuss philosophical issues related to personal identity preservation. For me things are clear enough, and I encourage you to read Hayworth's "Killed by bad philosophy - Why brain preservation followed by mind uploading is a cure for death" for a crystal clear analysis.

Chemical brain preservation has also simple operational advantages over cryopreservation. Since it does not require especially expensive storage facilities, it can be offered at a lower price (perhaps at a much lower price). Also, it may be already covered by existing laws and regulations. If chemical preservation can be legally considered as a form of embalming, as the Body Worlds art exhibit seems to indicate, then the preparation of patients, their transport to a storage facility abroad, and running a local storage facilities, are already permitted by the law in most countries.

From the Open Letter on Brain Preservation, which you should sign now: "We accept the current scientific consensus that our unique conscious self is generated by processes within our physical brain. Further, we accept that all the memories, skills, and personality traits that make us unique are hardwired into the physical and molecular connections among our brain’s hundred billion neurons... The structural basis of memory and personality -- the synaptic connectivity between neurons -- can be preserved essentially perfectly by today’s chemical fixation and plastic embedding techniques. Extrapolating from current technologies for the nano‐imaging of plastic embedded brain tissue, we believe that one day science will have advanced sufficiently to allow complete retrieval of memories from such a preserved brain."

Some references:

Ben Goertzel on KurzweilAI: "Ken Hayworth asked in his H+ Summit talk, “Can we extract a mind from a plastic-embedded brain?” His collaborator John Smart hit the same theme — and spent most of the conference sitting in the lobby at a table raising money to pay a summer intern to help with the research. With a PhD from USC and a post-doc at Harvard, Ken isn’t exactly an amateur — but he’s looking at his field with new eyes. Plastination of body tissues was developed for other purposes (did you see the beautiful Body Worlds art exhibit?), but it may well obsolete cryopreservation, posing a lower-cost and more effective way to preserve the minds of the deceased till the Singularity when they can potentially be reanimated."

George Dvorsky: "I've often thought that cryonics, the practice of storing tissue (namely the brain) in a vat of liquid nitrogen, may eventually come to be seen as a rather primitive and naive technique for preservation... brain plastination was recently given a considerable boost through the founding of the Brain Preservation Foundation. Launched by Accelerating Studies Foundation founder John Smart and Harvard neuroscientist Ken Hayworth, the BPF is seeking to facilitate the development of any technology that will effectively preserve the brain for eventual reanimation... As for mind uploading from a plastic embedded brain, Hayworth believes that's about 50 years off."

Fibur.ru: "Is there a way to preserve your brain, and thus your identity for the future? Traditionally, some people have turned to cryonics —basically freezing their brains and bodies in a vat of liquid nitrogen with the idea that in the future nanotechnology will be able to unthaw and revive them. At the summit, John Smart, president of the Acceleration Studies Foundation, announced the Brain Preservation Prize. (Disclosure: I made a $50 contribution to the foundation while at the summit.) Modeled on the X Prizes, the Brain Preservation Foundation wants to encourage researchers to develop techniques “capable of inexpensively and completely preserving an entire human brain for long-term storage with such fidelity that the structure of every neuronal process and every synaptic connection remains intact and traceable using today’s electron microscopic imaging techniques.”The idea is that the precise pattern of information in an individual’s brain constituting that person’s identity would be preserved and could be revived later by being uploaded into an advanced information technology network or perhaps a new body and brain. Although any technique could qualify for the prize, Smart evidently believes that a kind of plastination is the most likely way to go. People who attend the Body Worlds exhibition are familiar with one type of plastination that is used to preserve entire human bodies for display. One technique involves flooding a brain shortly after death with glutaraldehyde to fix proteins, followed by osmium tetroxide to stabilize lipids and other compounds. This process turns a brain into a black block of plastic that will last indefinitely. Smart was followed by Harvard researcher Kenneth Hayworth whose work focuses on using electron microscopy to delineate every synaptic connection from plastinated mouse brains. Plastination preserves both structure and molecular level information. He predicted that scientists would produce a synapse level atlas of an entire human brain over the next decade. “Can a mind be extracted from a plastic embedded brain?” ask Hayworth. “The answer is almost certainly yes.” When? In the next 50 years, predicted Hayworth."

H+ Magazine: "In the future, we might understand brain circuitry so well that such devices could be used to scan and “upload” an individual’s mind to any type of substrate (a new body, robot, or artificial environment). This Matrix-like immortality would be the ultimate backup of ourselves."

Aschwin de Wolf: "Mind uploading advocate Kenneth Hayworth has launched an interesting website devoted to the science of brain preservation. Of particular interest is his Proposal for a Brain Preservation Technology Prize (PDF). This document includes one of the most comprehensive discussions of chemopreservation as a strategy for personal survival."

Greg Jordan, in a 2008 article on Biostasis through chemopreservation: "Twenty years ago, Charles B. Olson published an article called “A Possible Cure for Death” in the journal Medical Hypotheses. In it, he favorably compares methods of chemical preservation to cryogenic preservation. Unfortunately, this article provoked no wide discussion or attempts at implementation... Chemical methods of preservation such as fixation are not only adequate, they have long been the gold standard for biologists studying the structure of cells and the brain... But what of reversibility? Olson dismisses the need for reversibility. The information in the brain can be retrieved and run on a different substrate - a new organic or machine brain.. For those who accept the method of resuscitation by scanning the brain and running it its processes on a different substrate (“mind uploading“), chemopreservation might present additional benefits. The chemopreserved brain, unlike the cryopreserved brain, is ideally suited to microscopic extraction of information."


  1. I've signed the petition, but I don't agree that chemo is superior to cryo for uploading purposes. Whatever best preserves structure is best for uploading purposes. It is also what is best for reanimation purposes. MNT can reverse chemical fixation to restore cellular viability almost as easily as it can restore viability to a vitrified brain. What it can't do is retrieve memories stored in ultrastructure that has dissolved into solution.

    Note that keeping things cool benefits dramatically from economies of scale. Use of a sufficiently large tank for enough people would bring the cost down dramatically. Tanks that will not need re-topped for 100 years or more are conceivable due to volumetric decrease in exposed surface area and increased insulation thickness available on large scales.

    But if a good chemo version of cryonics is developed, that's great. Brains stored that way will be more portable, can be stored in less centralized locations, and do not need to be protected against power failure.

  2. Hi Luke, of course I agree that whatever best preserves structure is best for uploading purposes. I am interested in the development and operational deployment of mind uploading, not so much in any specific candidate technology. But at this moment I am persuaded that chemical brain preservation may be the best option available with current technology.

  3. With upgraded equipment (including provisions for controlling post-cannula infusion pressures and infusion temperatures, as well as monitoring infusion rates or use of them as the independent variable with provisions for clipping any pressure peaks that might be encountered), as well as instruction in optimal cannulation (both femoral and carotid), and the use of chemofixation fluids vs. biopreservation oriented fluids (as advocated by Giulio on his blog here), this procedure could be performed by funeral directors vs. cryonics biopreservation teams, at vastly less expense for the persons whose brains were preserved.

    Post-infusion neuroseparation (amputation of the trunk) or (for those who found that to be esthetically unacceptable) brain removal by additional surgery could further reduce fees for those who elected this approach to brain preservation, by reducing the volume required for storage and the cost of modest temperature reduction (far short of LN2 levels) for reduction in any residual chemical reactions feared to be long-term sources of additional reduction of detail in final scanning of the brain.

  4. @Boundlesslife: cost is certainly an important factor for most of us, but I believe the real obstacles at this moment are legal regulations, and public acceptance.

  5. I'm real happy to see this development. A diverse array of techno-efforts aimed at techno-immortality is best path to success. I do think, however, there is a (natural) tendency to way overestimate the importance of copying our brain structure to copying our minds. I think our minds will be uploadable in good enough shape to satisfy most everyone by reconstructing them from information stored in software mindfiles such as diaries, videos, personality inventories, saved google voice conversations, chats, and chatbot conversations. The reconstruction process will be iteratively achieved with AI software designed for this purpose, dubbed mindware. Nor is it a situation of either-or. In a few decades most people will get totally comfortable with the fact that their personal identity transcends multiple substrates and instatiations. Will there be times when parts of an identity want to divorce each other? Sure. That's what lawyers are for. It'll be a growth industry.

  6. Hello Martine,

    Your ideas, to which I frequently refer as "Bainbridge-Rothblatt soft uploading", were discussed at the recent First and Second Online Workshop on Advancing Substrate-Independent Minds (ASIM). Here is what I wrote to the ASIM mailing list to summarize my position after the second workshop:

    I think Bainbridge-Rothblatt soft uploading is feasible in principle, but with two major caveats.

    1) We need data transfer rates from the brain to computer storage much faster, by several orders of magintude, to permit storing enough information to reconstruct a given person with a sufficient degree of identity preservation. Even so, the transfer of memories to storage would take decades and can only be practical with implanted, thought operated (or even automatically operated) devices. Here is where BMI and neural prosthesis come in.

    2) The memories (as usually I am using "memories" in a wide sense to include dreams, hopes, fears, emotional responses...) can only run on a substrate (hardware and system software) flexible and powerful enough to run them with a sufficient degree of identity preservation. The system software must include very powerful AI and a "me-program".

    I tend to guess that the me-programs may be essentially the same for different persons. It seems likely that in order to reconstruct a suitable me-program by reverse engineering we will need not only functional studies, but also and especially structural studies.

    If these guesses are correct, Bainbridge-Rothblatt "uploading patients" can be re-instantiated only _after_ most other parts of the ASIM program have been achieved.

    A couple of weeks ago I sent you an invitation to attend the public online ASIM lecture by Randal A. Koene on Realistic Routes to Substrate-Independent Minds, Teleplace, July 17, 10am PST. It would be great if you could find the time to attend and discuss your ideas in the Q/A after the talk.

  7. I'm really happy to see the interchanges developing on this blog, as above, and at this time only want to comment on Guilio's thoughts that:

    "I believe the real obstacles at this moment are legal regulations, and public acceptance." and "Since it does not require especially expensive storage facilities, it can be offered at a lower price (perhaps at a much lower price). Also, it may be already covered by existing laws and regulations. If chemical preservation can be legally considered as a form of embalming, as the Body Worlds art exhibit seems to indicate, then the preparation of patients, their transport to a storage facility abroad, and running a local storage facilities, are already permitted by the law in most countries."


    In 1989, at a conference hosted by the Cryonics Institute, I presented a paper, "New Directions in Cryonics", on line at http://www.lifepact.com/newdirections.htm which specifically proposes an intersection of a global kind between the mortuary industry and cryonics. At that time, it was not possible to envision that an even more compatible interface between cryonics and the funeral industry might come into being, as is now the case with chemopreservation, but it seems clear that brain preservation by chemostabilization is a practical solution to many of the obstacles (to brain preservation) now confronting biopreservation oriented cryonics.

    The above linked article also delves into possible memory losses and the need to in some cases replace large blocks of neural tissue with "generic" replicas borrowed from the brain maps of "donors" who the patient would designate as being acceptable sources of such maps. In this way also, that old article relates to the subjects under discussion here.

    "Sign me up for brain preservation by the methods discussed here, as soon as possible!"

    That’s the only thing I'd like to add, in closing out this comment!

    Oh, also, I’m in total agreement with the comments by Martine Rothblatt above, as to the primacy of data in the way of mindfiles, both in terms of the instantiation timeframe and quality of content, vs. reliance on brain map emulation as a primary approach or even an indispensable adjunct.

    If I were offered the option of the best possible cryonic preservation by current technology (vs. cremation or burial), or even perfected brain chemopreservation, at the cost of erasing my mindfiles, I’d decline it, even if that meant my biological person would otherwise be obliterated moments later and the residue distributed in the form of simple molecules across the space between here and the Orion Nebula.

    Boundless Life,

    Fred Chamberlain

  8. @Fred: your endorsement of novel and unconventional approaches to cryonics (in the very wide sense of personal identity preservation after biological death), such as chamical brain preservation or the even less conventional "soft uploading" approach of Martine, is very important coming from a person like you, with a long history of leadership in the cryonics movement.

    I am reading your essay again, and I think it is very relevant to the issues discussed today.

    Re public acceptance. I am really disgusted by some of the comments to the recent New York Times article on cryonics. What I find disgusting, is that so many intolerant bigots criticize RH for a personal choice which does not affect others. Don't like the idea of cryonics? Then look elsewhere, for fuck's sake. Nobody is going to force cryonics on you, so it is not your business.

    Some things should not have to be "accepted" by the society, but just be considered as private choices of the person(s) concerned.

    Yet, I see a growing nanny-state PCization of our society. We routinely accept unnecessary and stupid limitations of our personal freedom, _on matters which do not affect anyone else_, which we would never have accepted in the 60s or 70s. We are becoming a society of sheeple, and I wonder what we should do to change this.

  9. I Thks for the institute to take consideration through efforts to start thinking in the second person about my discoveries. It won't be that there is a one fit way to do functions like micro elementary brain expression.