Ken's Work

Eray Ozkural exa erayo at
Tue Apr 20 09:32:18 EST 2004

Hi Ken,

I think you remember that you sent your book AoK to me as well, and I
have read most of it, save for the last appendix or so.

Comments below...

"ken" <kpaulc@[remove]> wrote in message news:<tz6hc.1703$eZ5.1419 at>...
> "John H." <johnh at faraway.> wrote in message
> news:4084e995 at
> > "David B. Held" <dheld at> wrote in message
> > news:c5vtmg$42h$1 at
> > > "ken" <kpaulc@[remove]> wrote in message
> > > news:t9Kgc.2067$e4.1942 at
> > > > [...]
> > > > The problem is that I worked =Hard=, produced
> > > > AoK's synthesis of Neuroscience experiment, and
> > > > it was so far ahead of the field, that no one would
> > > > even Review it.
> > >
> > > Is that the only reason nobody would review it?
> >
> > An outright lie by Ken. M. Kirkcaldie went to
> > considerable trouble to review AoK. Having read
> > AoK myself I found MK's review to be fair and
> > reasonable, naturally Ken thought otherwise.
> Here is what transpired in bionet.neuroscience. If
> you read it, you'll see that all the 'reviewer' did was
> carry-through his own preconceptions, without ac-
> knowledging anything that's actually in AoK.
> AoK's Reification of "decussation", the single most-
> significant thing that's ever been done in Science, is
> not even mentioned, nor is anything else that was
> newly Reified in AoK [there's a list of 55 such newly-
> Reified things in AoK's Preface]. All the 'reviewer'
> does is with respect to "TD E/I-minimization", and
> not very much of rhat.

TD E/I minimization does makes some sense, and I could in fact point
out to you the mathematical basis of such optimization ideas. (So
there is in fact a basis to optimization!) However, there is some
serious lack of methodology in your work.

First of all, you keep talking about an information calculus, an
exposition of which I have not seen anywhere. Unless you state exactly
what the optimization procedure is, it cannot be evaluated. This is
the most important part. You are talking about two neural systems with
different but complementary functionality, and you give nice intuition
on how the wiring is designed to optimize behavior, etc. However, you
fail to accomplish this in an appropriate mathematical formalism and
computer models. Thus, we cannot review it!

Secondly, you do not seem to acknowledge the recent neuroscience
research. Some of the recent papers might actually support your
"computational" thesis, for instance about the role amygdala and
hippocampus... At any rate, you have to supplement your thesis with a
very thorough survey. Have a look at the neuroscience section of
cognews: I think you will find interesting new work there.
Needless to say neuroscientists have not been sleeping all these
years, and there are some radical new ideas (certainly not stolen from
you), some of which you can connect to your work. I also think you may
have to revise your work in light of experimental findings.

Third, you have to avoid marrying your sociological ideas with the
foundations of neuroscience. If you would like to do such, write it
separately, as an addendum to your book on neuroscience.

Fourth, you have to propose your theory of neuroscience as independent
from your ideas about physics. I think some of your neuroscience ideas
may be on the mark, while your claims about physics may be completely
wrong. At least that's a significant possibility. So, please avoid
connecting physics to neuroscience. Try to explain everything from the
synaptic level to the entire brain and forget the rest for a while.

I think it would take a lot of hard work for you take the above points
into consideration, but I also think it would be worth the effort. You
can convert AoK to html, do the above corrections and post it on your
web page, so that many more people can help you improve it. It's at
least an eccentric read!

Nevertheless, your book is not more "orbital" than some of the
"authoritative" books on the nature of mind. Take for instance,
Bennett and Hacker's stupid book with several hundred pages and a
hardcover, in prominent libraries; your work is assuredly light years
ahead of it. (I am now also imagining an AI book written by a
physicist, etc. Your ideas may be as important as several currently
published "popular science" or "philosophy" books about mind...)

Finally, while posting on USENET, please try to use written language
which you employed well in your book.

Best Regards,

Eray Ozkural

More information about the Neur-sci mailing list