I played my presentation (www.tedcloak.com) to a meeting of the Los Alamos,
NM, Unitarians last Sunday.
The first question after the presentation came from a retired physicist, who
wondered how all this PCT activity could take place in real time, given that
the action potential of a neuron requires about 100 milliseconds. I had to
admit I didn't know, and would get back with an answer when I had one.
Help!
Ted
[From Fred Nickols (2008.08.13.1702 MDT)]
I await with bated breath the answers of those who are in a position to respond on an informed basis. On my part, I'm guessing that the physicist's question is rooted in a linear, cause-and-effect view and the 100 millisecond issue, if correct and if an issue, might be negated by the simultaneously "live" nature of a closed-loop feedback-governed system. We'll see what the experts have to say.
--
Regards,
Fred Nickols
Managing Partner
Distance Consulting, LLC
nickols@att.net
www.nickols.us
"Assistance at A Distance"
-------------- Original message ----------------------
From: Ted Cloak <tcloak@UNM.EDU>
I played my presentation (www.tedcloak.com) to a meeting of the Los Alamos,
NM, Unitarians last Sunday.The first question after the presentation came from a retired physicist, who
wondered how all this PCT activity could take place in real time, given that
the action potential of a neuron requires about 100 milliseconds. I had to
admit I didn't know, and would get back with an answer when I had one.Help!
Ted
[From Bill Powers (2008.08.13.1726 MDT)]
I played my presentation (www.tedcloak.com) to a meeting of the Los Alamos,
NM, Unitarians last Sunday.The first question after the presentation came from a retired physicist, who
wondered how all this PCT activity could take place in real time, given that
the action potential of a neuron requires about 100 milliseconds. I had to
admit I didn't know, and would get back with an answer when I had one.
Relax. The action potential of a neuron can reset fast enough to cause firing at least 1000 times per second (i.e., every millisecond), and typical neural channels involve many parallel fibers all firing at once for an effective rate perhaps 50 times that fast.
Your friend is thinking of reaction times; in my tracking experiment in the new book I show that for visual-motor tracking the user's actual transport lag is about 130 milliseconds, or 8 frames of a 60-HZ vertical scan rate on a monitor screen. Program disk comes with the book. Control is continuous and smooth. Rumors to the contrary, control systems containing transport lags can easily be adjusted to produce continuous smooth control, with the effective bandwidth being a function of the lag time. Input and output operate concurrently and continuously, rather than alternating. Richard Kennaway, our house mathematician (thank God), shows mathematically how this works in an appendix for the new book. I just show it with a simulation.
Best,
Bill P.
At 04:00 PM 8/13/2008 -0600, Ted Cloak wrote:
[From Rick Marken (2008.08.13.2150)]
I played my presentation (www.tedcloak.com) to a meeting of the Los Alamos,
NM, Unitarians last Sunday.The first question after the presentation came from a retired physicist, who
wondered how all this PCT activity could take place in real time, given that
the action potential of a neuron requires about 100 milliseconds. I had to
admit I didn't know, and would get back with an answer when I had one.Help!
Ain't it awful to get asked a question during or after a talk only to
figure out a good comeback three days later. That's what happened to
me when I gave the talk at Ucla. I presume that the questioners (and
audience) figured that I had been had.
I don't know if I would have been able to come up with a good answer
in your situation. After thinking about this for some time in the
serenity of a beautiful California evening all I could come up with
was this: the questioner was probably thinking that the slowness of
neural conduction would delay compensatory action so much that it
becomes ineffective. Your braking response to the slowing of the car
in front of you comes too late for you to maintain your distance. The
is know as the "feedback is too slow" argument against the control
model of behavior. As Bill mentioned, it is a false argument but it
is so persistent that he (Bill) had to recruit a real mathematician
(Richard Kennaway) to prove that it's not true. Since you probably
don't want to go through a mathematical proof that the "feedback is
too slow" argument is spurious, even with a retired physicist, what
you might do is point out that a familiar control system, the
thermostat, control temperature quite precisely even though there is a
transport lag (feedback delay) created by the thermostat's perceptual
system: the bimetallic strip. It takes time for a change in
temperature (error) to be perceived by the thermostat; its the time it
takes for for the strip to contract enough to close the contact and
turn on the heater. This transport lag is "smoothed out" by the
gradual change in the air temperature caused by the heater and other
disturbances (external temperature) acting on it. The same kind of
smoothing occurs in the control loops in living organisms.
Best
Rick
On Wed, Aug 13, 2008 at 3:00 PM, Ted Cloak <tcloak@unm.edu> wrote:
--
Richard S. Marken PhD
rsmarken@gmail.com