An example of mad science or bad science journalism

Allen L. Barker alb at datafilter.com
Sun Mar 16 14:53:46 EST 2003


KP-PC wrote:
> 
> "Allen L. Barker" <alb at datafilter.com> wrote in message
> news:3E74266C.4C048A8A at datafilter.com...
> |
> |
> | Here's another article on the research:
> |    http://news.bbc.co.uk/1/hi/health/2843099.stm
> |
> | They say the research is to be presented to a neural
> | engineering conference in Capri, Italy, next week.  Maybe
> | get the proceedings.
> |
> | And rockets will never work in a vacuum, because there is
> | nothing to push against, etc.
> 
> With the brain, it's an entirely-different 'playing-field' because
> the brain is a living, ever-changing system architecture, which
> ever-changing-ness derives in neural-activation experience.
> 
> Part of this ever-changingness is with respect to the
> relatively-recently [a couple of years back] hippocampal
> 'neurogenesis' stuff. The hippocampus is continually being
> replensihed from a stem cell 'factory' - hard to duplicate in
> silicon.
> 
> So, even if a 'robotics' thing can be engineered, to replicate
> 'normal' biological function, it'd have to be able to not only
> duplicate the ever-changing-ness that derives in activation
> experience, and, somehow, duplicate all of the prior biological
> ever-changing-ness that derives in the prior neural activation
> experience.
> 
> It's the latter consideration that is the 'show-stopper - because,
> while the rest of the brain has done it the normal way, the
> 'robotics' thing, even if it can duplicate 'normal' biological
> function, has not.

I have not read their actual article, so I don't know the details
of what all they have done or claimed to have done.  It might be
an interesting article.  The brain is very adaptive.  Rather than being
a *drawback* to neural prosthesis, that is an advantage.  If you get
the closed-loop feedback right, the brain can adapt to many things.
Just as it can adapt to various brain insults, etc.

> So an overly-crude analogy with respect to introducing the 'robotics'
> thing is trying to repair a computer with parts from an entirely
> different generation of computers. This can be accomplished, but, at
> best, the result will be a kludge, requiring a lot of work-arounds,
> and the end result will have a lot of performance problems that a
> 'normal' system is not subject to.
> 
> The result, with respect to applications to the brain, will be a
> drastic altering of what it is that the person whose brain it is will
> experience.
> 
> That's complete 'failure'.

Not if the alternative is a person's death or complete disability.

> I stand on my Analysis, posted in my prior reply in this thread, that
> this 'robotics' stuff is being passed-off as something that it
> actually is not - that it is actually part of a much-larger program
> to build a 'machine' that attempts to 'implements the brain'.

I'm sure there are very active research programs there, both open
and classified.  Similarly for neural prosthesis.

> This's much more doable than is the 'machine'-biology hybrid.
> 
> I know because I designed the machine. My purpose was different,
> however. The machine 'design' was a 'tool' that I used while working
> to comprehend nervous system function. It just 'fell out' of the
> other work I was doing.
> 
> All of my comments, above, with respect to experience, still apply.
> 
> If folks 'ignore' that, then they're deciding to develop just another
> 'robotics' thing, and that 'thing' will have more in common with the
> 'machines' that were depicted in the "Terminator' movies than it does
> with Human nervous systems - which I'd not be surprised to find out
> is the whole premise of the larger project of which, it's my
> Analysis, this 'hippocampal' stuff is a part.
> 
> "DARPA" and ONR= "weapons systems"
> 
> "Neuroscience" = "understanding of Human nervous systems"
> 
> "And never the twain shall meet."
> 
> Some fancy stuff =can= be engineered and manufactured, and it'll have
> its usefulness, far outstripping contemporaneous 'computers', but
> it'll never 'clone' a Human in silicon. All it'll ever be is 'just'
> another 'tool'. And the closer it does get to 'the brain', the more
> it'll have to be dependent upon experience. The main advantage a
> machine would have with respect to such is that, once experiential
> databases are established, they can be 'migrated'
> [transformed-in-process] to succeeding generations of machines. But
> that's not an easy problem, and the hardest problem will remain the
> 'war' between mutually-exclusive experiences that is the main thing
> with respect to which evolutionary dynamics have struggled, over
> eons, with respect to their 'engineering' of biological brains.
> 
> There is, also, exceeding danger in such 'technology' - the
> abdication of Human Choice. I've seen a =lot= of such resulting from
> just the weak-kneed 'computers' that have already existed.
> More-advanced 'machines' will tend to augment such as folks, more and
> more, pass off the information-processing work that is necessary, in
> order to imbue their own brains with the necessary useful experience,
> to machines.
> 
> Machines are =tools=. To the degree that machines displace Human
> thought, Humans become less-than-Human.
> 
> I've witnessed a lot of such already, and I've not seen anything that
> leads me to believe that the folks engineering future machines are
> cognizant with respect to such, so I'm not optimistic with respect to
> such.
> 
> All I see is more of the 'blindly'-automated Prejudice that's Ravaged
> Humanity since the Beginning, itself being actively augmented.
> 
> K. P. Collins
> 
> | Peter F wrote:
> | >
> | > Go to www.newscientist.com and take a look at the article about:
> ~implanting
> | > a microchip to do the job of the hippocampus~.
> | >
> | > It made me check to see if it was April 1 already! Sadly, it was
> not.
> | > :-|
> | >
> | > P
> |
> | --
> | Mind Control: TT&P ==> http://www.datafilter.com/mc
> | Home page: http://www.datafilter.com/alb
> | Allen Barker

--
Mind Control: TT&P ==> http://www.datafilter.com/mc
Home page: http://www.datafilter.com/alb
Allen Barker



More information about the Neur-sci mailing list