[MURG] new Road Map draft

[cat13 at illrepute.org]

"king-yin yan" <y.k.y at lycos.com> writes:

> Firstly, some corrections:
> 
> Cortical neurons should be ~27 microns apart on
> average, not 35 (miscalculation). This is a rough
> figure anyways. Also, memory extraction should
> require cell-type identification. There are other
> assumptions to it as well. All files have been
> updated.
> 
> From: Ed Minchau <spider_boris at yahoo.com>
> >Consider that a new laptop computer of today is as
> >powerful as all of NASA's computer hardware, taken
> >together, circa 1969.  There are also parallel
> >advances in other related fields: nanotechnology,
> >mesoscale physics, materials science, artificial
> >intelligence, genetics, molecular simulation, and so
> >on.  
> 
> For those who want to be among the first uploaded,
> they'll have to accept a fridge-size if not room-size
> computer...

Wireless communication between the main processing machine
and a PDA-sized device attached to the belt, with wires
going from the belt-device to the skull.
This issue has already been addressed:
(oops, used to have a bunch of links on this, not sure
where I put them)
http://www.wired.com/wired/archive/10.09/vision.html
is the best I can find for now.

> >Is the best brain-computer interface microthin
> >Teflon-coated stainless steel probes poking through
> >the skull?  Can an equivalent be made using Carbon
> >nanotubes instead?  That way instead of a 75 [...]
> 
> I don't think any current probe is poking through the
> skull =)  Also, the current that a carbon nanotube
> can carry is measured with electron volts. This is
> MANY orders of magnitude smaller than neural signals.
> Thirdly we need to insulate the length of the probe
> from base to tip. Forthly we need to interface many
> brain areas simultaneously. Given these, the BCI
> problem is actually quite hard.

Excepting the fourth issue, these points have already
been addressed.  I wish I could find my links; search
for things like "neural prosthetics," "neural implants,"
"move cursor with brain," etc.  I don't have time now;
I try to look later.  I have a faint recollection of
glass-tipped microelectrodes.

Regarding the fourth issue - interfacing many brain areas
simultaneously - even if we could physically access many
areas right now, we wouldn't know what to do with the
data we'd get out of them.  I don't see the limited
physical access as really holding us back right now: we
can access one area, perfect and miniaturize our interface
with that, then hit the next area, and so forth.  I'm
suggesting that perhaps, by the time we know how to
interface with any one of those multiple areas, we will
have progressed so that we are able to get around the
physical limitations.