SV: Capacity of the brain

Tim Tyler tt at cryogen.com
Wed Sep 15 09:19:57 EST 1999


In comp.ai.alife Ken Collins <KPaulC at email.msn.com> wrote:
: Tim Tyler <tt at cryogen.com> wrote in message news:FI21B5.82F at bath.ac.uk...
:> Ken Collins <KPaulC at email.msn.com> wrote:

:> : another point Kursewile brought up was with respect to the implantation
:> : of 'helping' circuitry into brains. [...] while such implants are
:> : useful with respect to deficit conditions, any such stuff applied to
:> : intact nervous systems will, necessarily, decrease the
:> : information-processing power of those nervous systems...
:> : render them incomplete in significant ways.
:>
:> How so?  I extend the processing power of my bain on a daily basis -
:> by interfacing it to my computer, via visual and tactile feedback.
:> This enables me to do a number of things I would otherwise find
:> to be next-to impossible.

: trust me, every byte one pushes through one's machine has its price in terms
: of what one could've, otherwise, been... doesn't mean it's a 'waste'. just
: means what i said.

:> I don't see how the exact location of the processing hardware will
:> suddenly render this type of man-machine interface worthless.

: forgive me, please, that's because you don't comprehend how nervous systems
: process information.

: the main thing is explained in AoK, Ap9. at an overly-simplified 'level',
: the diminution of overall information-processing capacity of a 'normal'
: nervous system derives in the alteration of the neural topology
: necessarily-imposed by the 'implant'. this's because 'normal' nervous
: systems 'seek' to minimize circuit lengths... shove anything else in-there,
: and the nervous system, 1. doesn't know how to minimize circuit lengths
: within the thing that's shoved in-there, and 2. the physical presence of the
: thing that's shoved in-there prevents the minimization of circuit lengths at
: the place where it's shoved in-there, and this disruption of what's 'normal'
: has ramifications, widely, within the nervous system.

: that's why any 'normal' nervous system will always 'kick the butt' of any
: formerly-'normal' nervous system that has anything shoved in-there.

But this is hardly right.  A coprocessor to perform mathematical
operations could massively outperform neurons on the same task.  Basically
we can design the guts of such a co-processor relatively easily, while we
have no idea how to go about building one out of wet neurons in any
reasonable timescale.

Your point about signal distances does not apper to be relevant - a
maths co-processor could take up very little space.  /Perhaps/ the
space would be better occupied with neurons - but no-one knows how to go
about building a "neuronal" maths co-processor, while a silicon one can
at least be envisaged today.

Also, silicon substrates are inherently better than brain material at
performing large numbers of sequential calculations rapidly.  If you have
a problem that *requires* fast, serial computations, a neural net is
simply not the right sort of architecture to use - but it is practically
the /only/ type of circuiry you can build from neurons.

Kurzweil has machines supassing human brain material somewhere around the
middle of the 21st century.  I should think that - according to him - it
will simply be more efficient use of the available space to scoop out the
bulky, slow, neurons and replace them with modern nanotech crystalline
computational hardware ;-)

: if folks go at it on an evolutionary 'time' scale, will they 'get lucky'?

: nope, 'cause, during all that 'time', 'normal' nervous systems'll 'just' be
: getting better... there's a bit of Xeno in-there :-)

This doesn't necessarily follow either.  Normal nervous systems can only
evolve at the rate of the species that owns them.  By contrast simulated
evolution can go through hundreds of generations in a second on a single
computer for some tasks.
-- 
__________
 |im |yler  The Mandala Centre  http://www.mandala.co.uk/  tt at cryogen.com

The early worm gets the bird.



More information about the Neur-sci mailing list