Vogel says brain too complex to upload

From: Robin Hanson (rhanson@gmu.edu)
Date: Mon Apr 23 2001 - 11:13:05 MDT


I don't find this argument very persuasive, but FYI:

Here is the final report of an NSF conference on "Societal
Implications of Nanoscience and Nanotechnology, NSF":
http://itri.loyola.edu/nano/societalimpact/nanosi.pdf

On pages 146-147 one finds:

SOCIETAL IMPACTS OF NANOTECHNOLOGY IN EDUCATION AND MEDICINE
V. Vogel, University of Washington

...

Science fiction rather than reality: The popular press has often aired
heated discussions
by the author Ray Kurzweil and others about the idea that it will soon be
possible to scan
the human brain and essentially transfer its neural activity to a computer
designed to
simulate billions and billions of human neurons (Kurzweil 1999). This
fantastic thought
is based on a series of assumptions, some of which are reasonable
extrapolations of future
technological abilities. Others, however, completely neglect how little is
still known
about how the mind works. Imaging technology may indeed reach microscopic
resolution, which may reveal individual synaptic contacts between nerve
cells. If Moore’s
law can be extrapolated, computers will achieve the memory capacity and
computing
speed of the human brain by around the year 2020. Computer experts were
therefore
quick to postulate that copying the 3D neural circuitry of the human brain
would become
possible with these powerful computers and advanced imaging technologies.
Once this
is achieved, they claim, it will be possible to simulate first the brain
and its function and
eventually the state of the human mind, complete with its memories,
emotions and
creativity. But it is important to remember that these nightmarish
scenarios are put
forward without any real biological understanding of the brain. For
example, these
scenarios rely on the assumption that the brain is nothing more than a
hard-wired neural
network, and that knowledge of the 3D brain architecture would be
sufficient to assess its
functional states. This may be the case for nematodes — the worm C. Elegans
has a
nervous system consisting of 302 neurons whose connections are all known (White
1986). But the brains of higher vertebrates have fundamentally different system
architectures than computers. Furthermore, single neurons are highly
nonlinear systems.
Single neurons in the cerebrum can make more than ten thousand connections
to other
nerve cells. The picture gets even more complex with the recent findings
that higher
vertebrate brains show plasticity. Plasticity is the ability of a system to
change its
structure and/or function in response to injury, the environment and/or
other changing
conditions (for further readings see Jacobs et al. 2000; Malinow, Mainen,
and Hayashi
2000; Poldrack 2000; Simos et al. 2000; Tramontin and Brenowitz 2000).
Given this
complexity of the brain, a scan of the brain will not allow a read-out of
the human brain’s
mind nor its memory. A century ago, society was embroiled in an almost parallel
controversy as to whether the future was completely deterministic and
calculable based
on Newtonian mechanics. It took the discovery of quantum mechanics to defy
the notion
that our future is predictable.

Nanobots are often on top of the list of nanotechnological creations that
cause deep
concern to the public. Eric Drexler and followers postulate that it will
soon be possible to
create nanoscale, addressable robots that have the ability to move in
space, recognize the
environment and self-replicate (see e.g., Stix 1996 for a critical review).
Will it indeed be
possible to create another form of life at the nanoscale? When it comes to the
engineering of nanoscale machinery, nature is still far superior in its
ability to integrate
synergistically operating nanoscale systems of high complexity. Yet, even
nature has not
been able to engineer nanoscale creatures that combine all of the
above-mentioned
attributes of nanobots. Viruses are amazing nanoscale systems, but even
they do not have
the finesse of the hypothetical nanobots. Viruses are able to move and they
contain the
genetic blueprint of themselves, yet they are not capable of
self-replication. Since they
depend on the replication system and protein synthesis machinery of much larger
organisms, namely micro-scale cells, they do not meet the definition of a
self-replicating
system. While mankind is equipped with increasingly powerful tools to
manipulate living
systems, we are not at the verge of creating herds of synthetic
self-replicating nanobots
that will run out of control and threaten our lives. Future man-made
nanosystems will
certainly be able to perform a variety of functions, but a robot that is
proficient in all
three functions — movement in space, recognition of a chemically complex
environment
and self-replication — will remain the fabric of dreams.

Robin Hanson rhanson@gmu.edu http://hanson.gmu.edu
Asst. Prof. Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030-4444
703-993-2326 FAX: 703-993-2323



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:59:55 MDT