[From Jon Krispin (990511.1150 EDT]
Responding generally to the opposition thread:
Okay, I'll come out of the closet. I wrote the post to the LO list regarding Herrnstein's matching law that was reposted here by Fred Nickols.
I have been lurking on CSGnet for about 3 weeks now at the invitation of Fred (we first "met" on the LO list and have had some interaction offline). I had no idea that he was going to repost what I wrote to this list. I don't know that I would have been as bold on my own. Bygones at this point. I have been riveted to the resulting discussion and fascinated to see the response that it has generated.
By way of introduction, I have a PhD in Social and Organizational Psychology from Temple University. The majority of my formal education was in the S-R model, one with which I was never satisfied and have generally abandoned. Late in my graduate career, I was introduced to behavior analysis and its applications in organizations, and I have found this perspective to be very useful. In my estimation, the feedback loops for behavior formed by the consequences that are experienced provided a piece of the puzzle that was missing from traditional social and organizational psychology to which I had primarily been exposed.
I am not an experimental psychologist, and my own understanding of the intricacies of the behaviorist perspective and its research methods is still growing (my weakness in the area has been put on display for all to see). I have read Bill Powers book _Behavior: The control of perception_, also at the behest of Fred Nickols, and I am trying to understand more about PCT. BTW, I wanted my post to the LO list to accurately represent the behaviorist position and the matching law and, many thanks to Bruce Abbott, I now know that it was not entirely accurate - an error which I will note on the LO list.
My main interest in this list is to gain an understanding of what the relative advantages and disadvantages of PCT vs behavior analysis are in accounting for observed behavior. The discussion, proposed experiments, and thinking out loud by Bruce A., Bill P., and Rick M., have been right up this alley. I will readily admit that the expertise that I have observed on this list (both in PCT and behavior analysis) is far greater than my own. So I am content to read and learn, for the most part.
In terms of the interval, as opposed to ratio, schedule applicability for the matching law that Bruce A. has clarified - my post to the LO list was indeed in error. In my examples, I did incorrectly (in ignorance) use VR schedules, and the bone that Bill picked was entirely justified (this is the clarification that I need to make to the LO list). My last interaction with the matching law prior to the past several weeks was probably 3-4 years ago, and, obviously, I had forgotten some of the specifics. I was prompted to recall the matching law by another post on digestive models for change on the LO list. At any rate, It seems to me that Bill's understanding of the matching law and the schedules to which it is applicable may have similarly been in error. Maybe there is something to be learned in this for both of us.
A side comment to Bill on my mathematical ability. Obviously, I did not include parentheses consistently in the denominator of each of the equations as I should have. This was a mistake that I caught myself (immediately _after_ I pushed the "send" button), and one which I attribute to my own debate as to how to consistently format the equations so that nothing would be lost in the "reformatting" that would inevitably take place as the message was distributed to the list (to take a stab at framing it in PCT, I was controlling for a variable other than the parentheses when the error occurred). This formatting problem is also the reason for the varying cases of the subscripts for the variables. Contrasting the cases of the subscripts to differentiate them from the variable is a convention that has been used several times on the LO list. I don't think that it is nearly as distracting for members of that list given their familiarity with this formatting rule for the presenta!
ti!
on of equations as it may have been for you. Maybe it is not the best solution, but it has been functional on the LO list. To summarize, while I certainly would not claim mathematical wizardry for myself, I do claim some measure of competence.
Having said this, I am even more glad for the presence of Bruce Abbott on CSGnet. He has been presenting the matching law and the behaviorist perspective as I would have liked to, and, as a result, I am learning much about the perspective of PCT AND the matching law - far more than I had hoped for. The result has been a wrestling by various participants on this list with the implications of the matching law and how it may be accounted for by PCT. If my errors had been left to stand on their own, I doubt that the distraction that they present would have been overcome. I am still watching to see if the predictions of the matching law and PCT are in conflict and which perspective is disproven by the experiments and simulations that seem to be in the works. For the interaction around this topic to date, and for what is yet to come, I would like to thank all who have been participating.
Returning to lurking for now,
Jon Krispin