···
Richard Pfau (2013.11.29 11:51EST)
Regarding [From Rick Marken (2013.11.21.1730)] Attachment: Chapter 16 Experimental Methods: Summary
RM: If the organism is a control system then if the IV has an affect on the behavior (actions) of the system, then the IV is a disturbance to a controlled variable that is also affected by the actions of the organism.
RP: Isn’t this focusing a bit too much on the controlled variable and ignoring possible IV effects on other aspects of a control system including the Input Function, the Output Function, the Reference Signal and the Comparator?
RM: I focus on the effect of the IV on a controlled variable because that’s what the IV affects in a PCT model of the behavior in the kind of psychology experiments I’m thinking of; that is, behavioral experiments. In these experiments, the IV is always a variable that influences the state of a controlled variable (at least, so the model says; there are no controlled variables in the conventional model of behavior ).
RP: For example, can’t an IV such as a drug affect the Input Function and relatedly affect the behavior (actions) of the system?
RM: Yes, I suppose so. Whether that was what was happening would have to be determined by testing a model. My guess is that, in most cases, the effect of drugs on behavior results from the disturbing effects of the drug on some intrinsic controlled variables. Well, the hallucinogens might actual affect perceptual functions.
RP: Can’t a surgical procedure such as snipping neurons involved in perception affect the Input Function and the behavior (actions) of the system?
RM: Yes. But I’m pretty sure the people doing such experiments are aware of the fact that they are messing with input functions. And I’m not sure that I would put the effects of surgical procedures (or drugs, for that matter) that alter the physiological properties of the organism – input, output or comparison function – into the model as disturbances since there is not much the system can do (behaviorally) to protect controlled variables from their effects.
RP: Can’t the magical concept pf “priming” (i.e., sensitization or inhibition of neural circuits) affect the Input Function and the behavior (actions) of the system?
RM: Yes. For example, staying in the dark for a while apparently changes the sensitivity of the photoreceptors. But, again, I wouldn’t all this a disturbance; it’s just changing the characteristics of the input function, making easier (or harder, depending on the nature of the “priming”) to perceive a controlled variable.
RP: Can’t other IVs do likewise? (Including, perhaps, uncontrolled perceptions?)
RM: Sure. I think the important thing, in evaluating the results of psychological research, is to properly model the situation.
RP: The same for the Output Function/ Can’t IVs such as drugs, surgical procedures, priming (?), blood glucose levels, etc. affect the Output Function and the behavior (actions) of the system?
And perhaps the same for Reference Signals and the Comparator.
RM: Yes. But if you are dealing with a control system then in order to understand the consequences of what you to to any part of the system you have to properly model the organization of the system;s behavior; and in a control system a central feature of that organization is the controlled variable – what the system is controlling. And when you do model a control system, I would say that the name “disturbance” should be applied only to environmental variables that influence the state of a controlled variable. I wouldn’t call drug effects or surgical manipulations or even what you call “priming” (I’d call it adaptation, maybe) disturbances; they might have an effect on a controlled variable but those effects cannot be compensated for by the controller. Changing the input function surgically, for example, is like changing the sensor on a thermostat from a temperature to a humidity sensor. That’s not really a disturbance; it’s just giving the system a new perception to control; a new controlled variable. So there will have to be some minor changes to the organization or the controller – new ways of generating outputs per unit error – but the system itself can’t change the sensor back to temperature; it just has to control this new perception as best as it can.
But the far and away majority of the experiments done in psychology are behavioral experiments (even if the dependent variable now is often an fMRI or something like that) and what is always missing from these experiments is an understanding that the subjects is a control system who is controlling various aspects of their environment: controlled variables. So my focus on controlled variables is really nothing more than a focus on the fact that the subject subjects in psychological experiments should be modeled as closed -loop control systems (which implies the existence of controlled perceptual variables) or open loop systems (which implies that there is no controlled variable involved in the behavior.
RP: In short, an excessive focus on the controlled variable may be narrowing our thought too much, to the exclusion of effects that IVs may have on other components of a control system that affect the behavior (actions) of the system.
RM: Again, I don’t think it’s a focus problem; it’s a modeling problem. Researchers who know that they are dealing with a control system and know how to model the system properly then will know that their IV is not a disturbance to a controlled variable if it only affects the input, output or other physiological functions inside the organism. when I focus on controlled variables I am doing it for the sake of researchers who don’t know they are dealing with closed loop control systems. I’m trying to get them to understand that such systems control their perceptions; the perceptions they control are called controlled variables. Once a researcher gets this, they can properly model the organism/environment system and understand the role of their IV in their research. In the vast majority of behavioral research, the IV will be a disturbance to a controlled variable;
Best regards
Rick
–
Richard S. Marken PhD
www.mindreadings.com
The only thing that will redeem mankind is cooperation.
-- Bertrand Russell