Print

Print


A lot of interesting points are beginning to be made on this subject. To make sure we are all on the same page regarding a classical feedback control system model I'll provide a brief tutorial. Assume an input signal "x" and an output "y"


    + ********               *********
x ----* sum  *---------------* H(S)  *------- y
      ********               *********     |
         | -                               |
         |                   *********     |
         --------------------* G(S)  *------
                             *********

The transfer function from x to y is given by

H(s)/(1 + H(S)G(S))

y can be viewed as the position in space of lets say my finger.
x can be viewed as a signal representing where I want my finger
to be. The sum represents the error signal that says how far I
need to move a muscle to get to where I want to be based on where I am.

By using a second order function for H(S) I could produce a model to
mimick all the PD symptoms plus dystonia from over medication. This can be done
by changing phase and gain of the feedback function G(S). As long as I send
new values of x, in other words keep changing where I want my finger to be,
the x signal dominates and I don't have tremor. When I stop telling my finger to
move, I am now dependent upon my feedback control system to hold position.
Unfortunately, with certain gain and phase delay it will become a tuned circuit
which will continually over compensate for positional errors causing tremors.

We obviously have nerves in our body that tell the brain where our
limbs are. Whether this is by relative position or detection of state
of contraction of different muscles I don't know.

We obviously have nerves that control the muscles. But there is a big hole
between that part of H(S) that tells the muscle to contract and that part of
G(S) that tells the brain where the muscle or body part is.

Where is the summing node, is also a question. Is this a chemical reaction
between two neurotransmitters as Duvoisin's book would have you believe? Is it
the interaction of different types of receptors in the same pathway as might be
suggested by STEPHAN SCHWARTZ's post?

Some questions for you brain trust types to ponder over the weekend. I'm going camping.
It's a holiday weekend in the States. I look forward to having all these questions
answered when I return. Don't get a headache.

Phil G. 47/1  [log in to unmask]