[From Rick Marken (2002.06.04.1550)]
Bruce Nevin (2002.06.04 14:32 EST)
...
Could we create a spot on the web site for workshop descriptions?
You betcha. Describe away.
Best
Rick
PS. In the _Neuroscience_ section of "News of the Week" in _Science_, 31 May,
2002, vol 296, p 1587 there is a description of research done my Michael
Graziano and colleagues that is reported in the 30 May issue of _Neuron_. The
research involved stimulating neurons in the motor cortex for 1/2 sec rather
than 50 msec (1/20 sec) as had been done previously. The results with the 1/2
sec stimulation were dramatically different from those obtained with 1/20 sec
stimulation. With 1/20 sec stimulation you get brief muscle twitches; with 1/2
second you get "natural movements" like moving the hand to the face. The
results are readily understood in terms of PCT if we imagine that what is being
"stimulated" is not a motor signal but a _reference_ signal. It takes time to
produce the perception indicated by the reference signal. So the longer the
reference signal is "on", the closer the system can come to producing the
intended perception. I think this would be _very_ easy to model. It looks like
another beautiful (and unintended) confirmation of PCT to me!
···
--
Richard S. Marken, Ph.D.
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org
i.kurtzer (2002.06.04.2130)
[From Rick Marken (2002.06.04.1550)]
PS. In the _Neuroscience_ section of "News of the Week" in
_Science_, 31 May,
2002, vol 296, p 1587 there is a description of research done my
Michael
Graziano and colleagues that is reported in the 30 May issue of
_Neuron_. The
research involved stimulating neurons in the motor cortex for 1/2
sec rather
than 50 msec (1/20 sec) as had been done previously. The results
with the 1/2
sec stimulation were dramatically different from those obtained
with 1/20 sec
stimulation. With 1/20 sec stimulation you get brief muscle
twitches; with 1/2
second you get "natural movements" like moving the hand to the
face. The
results are readily understood in terms of PCT if we imagine that
what is being
"stimulated" is not a motor signal but a _reference_ signal. It
takes time to
produce the perception indicated by the reference signal. So the
longer the
reference signal is "on", the closer the system can come to
producing the
intended perception. I think this would be _very_ easy to model.
It looks like
another beautiful (and unintended) confirmation of PCT to me!
I think you're seeing confirmations where there are alot of possibilities.
Any system that has damping will take time to reach a steady state,
provided it is stable. I do think there are plenty of poetential links to
PCT as they are finding more and more "perceptual" responses in "motor
cortex" and "motor" responses in "posterior parietal/sensory" cortex.
See "The organization of the cortical motor system: new concepts"
Rizzolatti, Luppino, and Matelli
Electrocephalography and Clinical Neurophysiology, 106, 1998, pp283-296.
The idea of a hierachy of loops seems lost on everyone. They generally
argue over a hierarchy vs loops.
What I find amazing is the obscuity of the stimulus procedure. I have been
in the field for awhile and have read plenty of papers and never noticed
it. Maybe I read too quickly, thats likely. But its one of many facts
that are known to all the players but noone outside. Thanks for the
reference.
Isaac
[From Rick Marken (2002.06.04.2000)]
i.kurtzer (2002.06.04.2130)
I think you're seeing confirmations where there are alot of possibilities.
Any system that has damping will take time to reach a steady state,
provided it is stable.
Did I forget to mention that stimulation of the efferent neuron causes the
organism to produce the same _result_ (such as touching the face), not the
same actions? Of course, this could still be the behavior of a damped, causal
system (like a pendulum). We'd have to look at the study to see if they
introduced disturbances _during_ the stimulation period. I doubt it. But if
they did, the number of possible explanations that don't involved control of
perception goes pretty much to zero.
Best
Rick
···
---
Richard S. Marken
MindReadings.com
marken@mindreadings.com
310 474-0313
i.kurtzer (2002.06.04.2300)
[From Rick Marken (2002.06.04.2000)]
> i.kurtzer (2002.06.04.2130)
> I think you're seeing confirmations where there are alot of
possibilities.
> Any system that has damping will take time to reach a steady
state,
> provided it is stable.
Did I forget to mention that stimulation of the efferent neuron
causes the
organism to produce the same _result_ (such as touching the face),
not the
same actions? Of course, this could still be the behavior of a
damped, causal
system (like a pendulum). We'd have to look at the study to see if
they
introduced disturbances _during_ the stimulation period. I doubt
it. But if
they did, the number of possible explanations that don't involved
control of
perception goes pretty much to zero.
I know, I have the paper in front of me. They did introduce an obstacle
between the hand and mouth, "the hand did not move around the obstacle but
hit it and was stopped". They don't say whether the object was totally
immobile or merely a substantial load. There are 1000 variations that they
should and they will likely do half.
Whether its the half that will be more informative about "control of
perception" is a hit-or-miss proposition.
i.
Best
Rick
---
Richard S. Marken
MindReadings.com
marken@mindreadings.com
310 474-0313
Isaac Kurtzer
Ashton Graybiel Lab of Spatial Orientation
Brandeis University
[From Bruce Nevin (2002.06.05 12:02 EDT)]
···
At 09:03 AM 6/5/2002 -0500, Richard Marken wrote:
Rick Marken (2002.06.05.0900)--
Graziano, M., Taylor, C and Moore, T. (2002) Complex movements evoked by
microstimulation of precentral cortex. _Neuron_, v. 34, 841-851
I can forward the paper to whoever would like it. It's in PDF format.
Can you post a URL whence we can download the PDF?
/Bruce
Thanks for the PDF, Rick.
I agree that this seems to be quite striking confirmation. Can a followup be submitted as a letter?
I especially enjoyed the part of the prolog where they are puzzled why the expected cognitive map of the body appears to be disorganized. Once again, what we expect to see gets in the way of seeing what to expect.
/Bruce
···
At 01:04 PM 6/5/2002 -0400, Bruce Nevin wrote:
[From Bruce Nevin (2002.06.05 14:11 EDT)]
At 09:03 AM 6/5/2002 -0500, Richard Marken wrote:
Rick Marken (2002.06.05.0900)--
Graziano, M., Taylor, C and Moore, T. (2002) Complex movements evoked by
microstimulation of precentral cortex. _Neuron_, v. 34, 841-851
[From Bruce Nevin (2002.06.05 22:17 EST)]
If you are planning on presenting a tutorial in modeling and programming please let us know what software or other equipment we need in order to participate.
Another idea for a tutorial: PCT experimental design. What is required to gather data to be modeled? Most of us don't have access to laboratories with rats, or the like. Can we nonetheless make serious contributions? Methodological guidance may help us to supplant theory-to-behavior inferences with data-to-model deductions, and shift our discourse from debate to demonstration.
/Bruce
Bruce Nevin (2002.06.04 14:32 EST)--
···
At 02:31 PM 6/4/2002 -0400, I wrote:
Last November we talked a bit about the need for some pre-arranged tutorials in modeling and programming which could as a byproduct yield materials for Dick's intro text book, or perhaps a programming workbook associated with it.
We seemed to have consensus about the following set of tools:
Excel
Vensim
SIMPCT (older version) if recoverable
Delphi 6.0 personal edition (Pascal)
Which we use, and how much we use each, depends upon people proposing workshops, putting together workshop materials, and letting the rest of us know about them in such a way that we go get the required tools and sign up.
[From Bruce Nevin (2002.06.04 14:32 EST)]
Last November we talked a bit about the need for some pre-arranged tutorials in modeling and programming which could as a byproduct yield materials for Dick's intro text book, or perhaps a programming workbook associated with it.
We seemed to have consensus about the following set of tools:
Excel
Vensim
SIMPCT (older version) if recoverable
Delphi 6.0 personal edition (Pascal)
Which we use, and how much we use each, depends upon people proposing workshops, putting together workshop materials, and letting the rest of us know about them in such a way that we go get the required tools and sign up.
Could we create a spot on the web site for workshop descriptions?
/Bruce
[From Rick Marken (2002.06.05.0900)]
i.kurtzer (2002.06.04.2300)
I know, I have the paper in front of me. They did introduce an obstacle
between the hand and mouth, "the hand did not move around the obstacle but
hit it and was stopped". They don't say whether the object was totally
immobile or merely a substantial load. There are 1000 variations that they
should and they will likely do half.
Whether its the half that will be more informative about "control of
perception" is a hit-or-miss proposition.
I now have the article in front of me, too, and I think it will be a PCT
classic. For those of you writing books on PCT, the reference is:
Graziano, M., Taylor, C and Moore, T. (2002) Complex movements evoked by
microstimulation of precentral cortex. _Neuron_, v. 34, 841-851
The results shown in Figures 1 and 2 are worth the price of admission. I
can't see any easy way to explain these results except using control theory.
The monkey is working against a variable disturbance (the different angles of
the force of gravity relative to the starting positions of the hand) to
produce a consistent result. I could produce a control model of these results
in a day. I think it would be very difficult to produce a caused output model
(the one the authors seem to advocate) of these results, if it's possible at
all. The fact that an obstacle stops movement is not surprising; a higher
order system would have to reset the reference for the system controlling
movement to the face; and that reference can't be changed because it is coming
from the experimenter's electrode.
This paper is not only strong evidence that cortical efferents are references
for perceptual inputs, it is also strong evidence for Powers' "neural current"
model of reference _and_, hence, perceptual signals.
I can forward the paper to whoever would like it. It's in PDF format.
Best regards
Rick
···
---
Richard S. Marken, Ph.D.
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org