Re: Solo Brain: was[nanotech] the core issue

From: Anders Sandberg (
Date: Mon Apr 03 2000 - 06:19:48 MDT

I agree with Charlie's somewhat pessimistic assessment of keeping
isolated brains alive - it is quite hard. Some successful experiments
have been done based on attaching the severed head to the circulatory
system of the body through tubes, but that is really cheating and
obviously not a long-term solution.

Charlie Stross <> writes:

> The nutrient requirements of a whole body are not exactly simple. I was
> once involved in the preparation of TPN (total parenteral nutrition)
> bags for patients who were too ill to digest food; TPN is a rich,
> creamy emulsion that can be perfused through a vein and contains about
> the minimum necessary to keep you alive -- vitamins, ions, glucose,
> trilgycerides, and a load of other stuff. It tends to be unstable --
> use within six hours -- and has to be prepared in a clean room with
> aseptic technique (you really _don't_ want to inject bacteria into
> someone on TPN!).

Still, it has become good enough to keep people alive for decades by
now. A brain might be somewhat simpler, since its nutritional
requirements are a bit simpler than the rest of the body, but TPN is
still hard.

> And the
> liver problem is more serious; keep recirculating blood through your
> disembodied brain and it'll poison itself with its own waste metabolites
> unless you've got a whole bunch of liver cells filtering the crap out of
> it. The nearest thing to an artificial liver we're likely to see in the
> next decade will be an external support column with a bunch of hepatocytes
> (hopefully differentiated from clonal embryonic stem cells) growing in it.

This has apparently been achieved, although I don't know how well it worked.

> Next, you need oxygenation and perfusion and circulation. This is
> distinctly non-trivial, without a heart and lungs; we have machines to
> do it in operating theatres, but they tend to knock the hell out of the
> erythrocytes and need a constant dribble of anticoagulants to keep from
> clogging up. And then there's haematopoeisis -- all your erythrocytes (red
> blood cells) die by 14 days, so you need to produce a constant supply of
> fresh blood. So either you need bone marrow, or your brain is going to
> be queueing up at the blood bank right behind Dracula.

What an image! A brain in a vat with vampire fangs... :-)

> The brain has sod-all immune response. So you need to keep the whole
> thing in a clean-room environment like unto a chip fab line -- operating
> theatres are _much_ too dirty for this sort of job.

It has some immune response from the microglia, but honestly they are
more trouble than use - their response to infections leaves much to be

> Finally, there's sensory hook-up. We are nowhere near being able to
> give someone artificial vision (at any level much better than being
> able to distinguish night from day). Nor do we know how to hook up the
> various cranial nerves to a vocoder. Your disembodied brain is, in fact,
> going to be in sensory deprivation hell -- until it succumbs to a passing
> antibiotic-resistant infection, which will happen sooner rather than later
> (because of all the crap we're having to ferry in to keep it alive).

On the other hand, keeping it unconscious is quite doable and useful
if the brain-in-a-vat stage is just temporary (like when trying to
find a body donor).

> The best life-support system for a human brain today is still a working
> human body. And I don't see this situation changing in less than one
> decade. Without spinal regeneration and full-body cloning, or much better
> prostheses with working neural control interfaces and sensory feedback, I
> can't see any medical demand for the Donovan's Brain treatment. (And I
> expect those prostheses not to show up -- it's looking right now as if
> it's a race between bionics and regeneration, and regeneration is
> winning.)

Regeneration can still likely be used to improve bionic interfaces by
having cultured cells attached to a chip connecting to the brain; this
will likely be less common and more expensive if everybody else is
using regeneration for the patients, but it would still be important
for some forms of paralysis.

Anders Sandberg                                      Towards Ascension!                  
GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y

This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:09:00 MDT