CIA-RDP96-00791R000200190067-4

Approved For Release 2003/09/16 : CIA-RDP96-00791R000200190067-4
Subject: Revised Notes from last e-mailing
SG11
Help!
I am a victim of too quick on the UNIX mouse button and mailed you
my thoughts before I proof read them. Feeek ... they were full of too fast
typing mistakes and outright misspellings. I am very sorry. Please junk that
last mail and use the corrected version below.
[I should slow down in
my typing 120 wpm is not good! !! ( :- ) ]
Revised notes follow:
Hello. As I begin going through an extraordinary mound of material a
thought occurred to me. I do not know exactly what your time constraints
are, but of the amount allocated to you by the Congressional directive, you
might consider contracting to us for
of it for the following 7-months
work beginning ASAP. (This is NOT an formal or informal proposal -- just an
SG1D
idea to help you with what appears to be a very difficult job. )
SG1D
For 6-months, you and I would work closely together to 'data-reduce' the
project into a manageable form for proper review by the NAS. I would suggest
the following approach to spark our discussion on the 28th.
I. Applications
involves judgments which have por
Given access to the Ft. Meade data and protocols, we would amass a 'success'
book, a 'failure' book and track repeat customer hits. (Any other critical
-
analyses
How much mat'& are we dealing with ? Could be done in componer;
that we can think of would also be added. ) Giving the NAS anything short
of a 'reduced' data set would be impossible to interpret by any honest review
E
scientists. All this might include positive and negative testimonials that
exist. Also, we might include selected examples -- I am generally not in
favor of showing gee-whizz examples because they can significantly mislead,
Approved For Release 2003/09/16 : CIA-RDP96-00791R000200190067-4
NÃO POL

Approved For Release 2003/09/16 : CIA-RDP96-00791R000200190067-4
but perhaps in this case it might be instructive to illustrate specific
points.
[ I define application-oriented research as that which is primarily aimed
at improving the AC output. Basic research is that which is primarily
aimed at understanding the mechanisms. ]
II. Application-Oriented Research
We will provide a single (readable) document that describes a meta-analysis
(psychology jargon for quantitative review) of all the government-sponsored
research in this domain. That would include selecting receivers
(i.e., human subjects), training them, assessing (analyzing) their output,
and
protocol issues (e.g., what are the procedural, physical, and psychological
circumstances for optimal performance.) Such an analysis would include
what is "known" and why, what is suspected, and what seems not to be true.
It might also include what would be done in the future, given that the
program continues.
Stil helpful to organice
into these areas, regardless
III. Basic Research
7
JOE?
We will provide a single document that describes a meta-analysis of all the
government-sponsored basic research. That includes issues like physical
parameters (e.g., grad(Shannon Entropy), brain wave activity, and
information transmission characteristics) . Knowns, unknowns, and future
directions would also be included.
IV. A number of 5-year plans have been written. We would up-date the most
recent one as a template on how to proceed, IF it is decided to do so. This
also could include a list of knowns, unknowns, and speculations and their
justifications.
who wrote there ?
V. There have been a number of government-sponsored reviews of this
work. We would provide to you critical analysis of those reviews and,
where appropriate, the rebuttals, and hard copy of the original pertinent
publications. That would include reviews by: NRC, OTA, Army, Two OSD, and
DIA.
Other - 7
PPIs.
VI. We will provide access to an august group of national/international class
scientists and government representatives who would be able to provide
independent views to the NAS panel.
VII. Other respected scientists have published in main-stream journals
showing the results of their analysis and experiments which suggest an
information-transfer-anomaly (we call Anomalous Cognition)
We should
Approved For Release 2003/09/16 : CIA-RDP96-00791R000200190067-4

. provide a review of this Literature.
Approved.For Release 2003/09/16 : CIA-REP96-00791R000200190067-4
ong
abelline!
1
At the end of a 6-month period, we would have a manageable amount of
material that could then be properly assessed by the NAS during the last
month of the contract (i.e., through the month of September). Depending upon
the output of such a critical review, your place can decide what, if anything,
you might want to do in FY 1996.
SG1D
Beginning March 15 and Ending 30 September,
is required.
That includes me and a technical assistant full time, part time administrative
assistant, and one consultant
receiver who has personal memory of the applications back to 1979. In fact
he was responsible for most of the data to 1983 (or whenever he retired) .
It also includes travel and the usual fees and SAIC bean-counting.
We would include contributing, as required, to the NAS investigation of this
material. My point is not to hide anything; rather, it is to organize a
20 year research and applications program into some manageable form. Otherwise
the NAS might be on your case ( :- ) .
Again, Andy, this is just a thought where I think we can contribute given that
the timing works out. The directive as I understand it would be easily
satisfied with such an approach. At the moment, I cannot think of a
better way to provide an honest assessment of the $20M-20 year program.
Let me know what you think of this idea. . . any of it can be changed or
ignored as the case may be.
Thanks,
Fd
Edwin C. May, Ph. D.
Director, Cognitive Sciences Laboratory
Science Applications International Corporation
330 Cowper Street, Suite 200
Palo Alto, California 94301 USA
Voice:
(415) 327-2007
e-mail:
may@hildegard.saicmp.com
Approved For Release 2003/09/16 : CIA-RDP96-00791R000200190067-4