[From Fred Nickols (2006.06.29.1434 EDT)] --
As mentioned earlier, I've been re-reading B:CP and I have another question or two.
In the Appendix, page 274, the diagram there shows the disturbance function feeding directly to the input function. I can see how that enables a perception of the disturbance itself but I was expecting to see the disturbance function affecting the environment function. So, after some thought, I've concluded that the diagram is correct in that it allows for a perception of the disturbance itself and, further, any effects of that disturbance on the environment function would simply take the form of a change in or to environment function. A perception of the disturbance as "causing" the change in the environment function is a different matter. So, here's my question: Do I have that correct?
Second question: On page 194, there is a discussion of "aversive reinforcement" under the heading of punishment. Is "aversive reinforcement" meant to be the same as what behaviorists would term "negative reinforcement"? Or was the term "aversive reinforcement" coined just to refer to punishment in different terms. I ask because punishment and negative reinforcement are two very different things.
···
--
Regards,
Fred Nickols
Senior Consultant
Distance Consulting
nickols@att.net
www.nickols.us
"Assistance at A Distance"
[From Bill Powers (2006.06.29.1255 MDT)]
Fred Nickols (2006.06.29.1434 EDT) --
In the Appendix, page 274, the diagram there shows the disturbance function feeding directly to the input function. I can see how that enables a perception of the disturbance itself but I was expecting to see the disturbance function affecting the environment function.
You gave me quite an anxious moment there. The disturbance feeding directly into the input function? I'm glad to say that Fig. A-1, page 274, shows the disturbing quantity, through the disturbance function,directly affecting the input quantity, qi. The input quantity is a physical variable in the environment outside the control system. The control system senses the state of the input quantity, not the state of the disturbance ("d" in the diagram). The state of the input quantity is determined by the sum of two influences, one from the disturbing variable d and the other from the output quantity qo. So there's no way for the control system to tell how much of qi comes from the disturbing quantity and how much comes from the output quantity. Nor does it need to know.
The perceptual signal represents the state of the input quantity qi, not the disturbance or the output quantity.
Second question: On page 194, there is a discussion of "aversive reinforcement" under the heading of punishment. Is "aversive reinforcement" meant to be the same as what behaviorists would term "negative reinforcement"? Or was the term "aversive reinforcement" coined just to refer to punishment in different terms.
It's discussed under the sub-heading "Punishment", so you can assume it was meant to refer to punishment. I knew even then that negative reinforcement was the removal or prevention of an aversive stimulus.
Best,
Bill P.
[From Fred Nickols (2006.06.29.1502 EDT)] --
Thanks for the clarification, Bill. Makes sense to me.
···
--
Regards,
Fred Nickols
Senior Consultant
Distance Consulting
nickols@att.net
www.nickols.us
"Assistance at A Distance"
-------------- Original message ----------------------
From: Bill Powers <powers_w@FRONTIER.NET>
[From Bill Powers (2006.06.29.1255 MDT)]
Fred Nickols (2006.06.29.1434 EDT) --
>In the Appendix, page 274, the diagram there shows the disturbance
>function feeding directly to the input function. I can see how that
>enables a perception of the disturbance itself but I was expecting
>to see the disturbance function affecting the environment function.
You gave me quite an anxious moment there. The disturbance feeding
directly into the input function? I'm glad to say that Fig. A-1, page
274, shows the disturbing quantity, through the disturbance
function,directly affecting the input quantity, qi. The input
quantity is a physical variable in the environment outside the
control system. The control system senses the state of the input
quantity, not the state of the disturbance ("d" in the diagram). The
state of the input quantity is determined by the sum of two
influences, one from the disturbing variable d and the other from the
output quantity qo. So there's no way for the control system to tell
how much of qi comes from the disturbing quantity and how much comes
from the output quantity. Nor does it need to know.
The perceptual signal represents the state of the input quantity qi,
not the disturbance or the output quantity.
>Second question: On page 194, there is a discussion of "aversive
>reinforcement" under the heading of punishment. Is "aversive
>reinforcement" meant to be the same as what behaviorists would term
>"negative reinforcement"? Or was the term "aversive reinforcement"
>coined just to refer to punishment in different terms.
It's discussed under the sub-heading "Punishment", so you can assume
it was meant to refer to punishment. I knew even then that negative
reinforcement was the removal or prevention of an aversive stimulus.
Best,
Bill P.