Re: Dissociation and emotion

Hi Fred:

Thanks for the initial compliment and sorry to have horrified you. Given that there is no actual threat of harm here, I trust you won't develop PTSD and that you will recover from the shock.

All kidding aside, your post suggests to me that we don't necessarily share certain aspects of our meta-theories, or views about approaching psychology. So let me respond to you not in attempt to convicne you I'm right, but to let you know where I'm comming from.

First and foremost, I am a methodological behaviorist. I am fundamentally interested in the description, control, and explanation of behavior.

Second, I am a material monist, which means that in principle, I believe that everything psychological fundamentally can be reduced down to something biological, which in principle can be reduced down to physics.

However, I am also a strong proponent of the idea that, just because we should in principle be able to reduce things to biology doesn't mean that it is always most useful to do so.

Third, I have been very strongly influenced by the philosopher of science Imere Lakatos. Lakatos built on Popper's notion that, logically, we can never prove a theory right, we can only prove a theory wrong. This is because the form of reasoning used to confirm a theory is in fact a logical falicy. It follows the form: My theory X predicts phenomenon Y. I observe phenomenon Y. Therefore I conclude theory X is true. Or in simpler terms, If X then Y, Y, therefore X. While this form of reasoning is quite familiar to us, it is in fact illogical or in valid. It is so well known to philosophers and logicians that it has a special name: Affirming the consequent. It is in valid or illogical because, even though it sometimes yields and accurate conclusion, it does not always. Let me give two examples to illustrat this:

Example 1:
P1. If John Kennedy was assasinated, then he is dead.
P2. John Kennedy is dead.
C: Therefore he was assisinated.
In this case, P1 is true and P2 is true. It also happens that, in this case, the conclusion is true. Contrast this with a variation in which I've kept the form of the reasoning, but changed P1 just a tad.

Expample 2:
P1: If Elvis Presly was assasinated, then he is dead.
P2: Elvis is dead (for those of you who believe he's alive, just humor on this on, OK?)
C: Therefore he was assasinated.
Again, both P1 and P2 are true. If Elvis had been assassinated, he would in fact be dead. This has to do with the definintion of being assassinated. And Elvis is dead. However, the conclusion that Elvis was assasinated is not true. Rather, he died of other causes. Thus, we can't use observations to prove a theory true.

Popper noted that we can use observations to prove a theory wrong. This follows a form of reasoning called modus tollens and goes like this: If theory A is true, when we should see phenomenon B. We don't see phenomneon B. Therefore, A is false. This syllogism is formally valid. But, there is a practical problem with this. When we develop a theory and put it to the test, we must invoke lots of other assumptions regarding how we define, manipulate, and measure the variables of interest. Thus, we actually have a more complex syllogism, the follows the form:
1. If Theory A and auxiallary hypotheses h1 ...hn are true, then we should observe phenomenon B.
2. We fail to observe phenomenon B.
Therefore, the conjunction of Theory A and auxiallary hypotheses H1...Hn is wrong. But, we don't know which aspect(s) of the this conjunction is wrong. It could be the the theory is right, but we are measureing the constructs wrong. Or it could be we measuring things right and the theory is wrong. Or it could be that both the theory is wrong and we are measuring it wrong. We don't know where the fault lies.

So, Lakatos basically notes that given this insurmountable logical pickle (we can't logically prove a theory right, and we can't practically prove a theory wrong), we have to evaluate progress in science not from the perspective of right or wrong, but by some other standard. He proposes this standard should be one of usefulness. A good theory makes predictions about things not yet observed and, at least once and while, turns out to have been correct in it's predicitons. When it is wrong, we either alter or abandon the theory.

So, for me, I assume all theories are wrong. What I want is a theory that is helpful to me in doing my clinical work and my research. From this point of view, are no a priori reasons to prefer a biological level of theorizing to a psychological level of theorizing. For what I do, I find a psychological level more useful than a biological level. While I fully realize that I am in fact changing patient's brains when I do therapy with them, the utility of a psychological theory is not dependent on being "proven" at the biological level. Thus, exposure therapy was effective and changed brains way before we had the technology to show this. And people should have been convinced of the efficacy of exposure and response prevention for OCD based on the behavioral data that has existed for quite some time. The validity of exposure therapy is not predicated on the ability to show that it causes changes in the brain. I am in agreement with Skinner on this one point: I'm interested in the impact of the environment and learing experiences on behavior. It's nice if biological theory data converges with behavioral theory, but the reality of functional relationships between the certain experiences and behavior does not have to wait for validation from the biologists. Just like the reality of biological science doesn't have to wait for the physicists to explain everything.

So, given that background, I don't much follow neuroscience research because I don't find their insights to be helpfull with the specific things that I'm interested in studying. What I rely on are pscychological theories of learning, primarily from the associative learning tradition (Pavlovian and instrumental learning). From that level of analysis, it is perfectly defensable to promote a theory that says, if I train a rat by giving it a series of CS-US pairings, the rat learns to, in some sense of the word, expect the US will follow the CS. The idea that Pavlovian conditioning is the result of a simple transfer of control of a reflex is simply not justifiable these days. There are tons of data that show this theory (at least in it's current form, given our current measurment practices) cannot be true.

So, this is a long way of saying the following:

1. All theories, including the theory of emotional processing which I personally find to be of use are wrong.

2. I none-the-less use emotional processing theory because I find it to be useful in guiding my clinical work and my research.

3. I don't worry too much as to whether or not psychological theories are currently supported by or refuted by biological data because at present, I don't find biological theories to be of much use for the things I'm interested in studying. That's not to say that other people won't, can't, or don't. Also, I'm not permanently married to emotional processing theory or to psycholoigical theories in general. If you can show me a biological theory that will make clear, testable predictions about how to improve the efficacy of psychological treatments for anxiety disorders, and show that these predictions are unique to biological theory and that psychological theories are either silent on the issue or make the opposite prediction, then I'm interested. Because then we can do the relevent experiment, the results of which will help us to advance knowledge which is what science is all about.

There you have it, my defense of using psychological theories of learning for my work, which may or may not include such concepts as beliefs and expectancies and distraction.

Hope you've recovered from the shock!

Replies:

Re: Dissociation and emotion, by Fred H., 02/09/03

| Behavior OnLine Home Page | Disclaimer |

Copyright © 1996-2004 Behavior OnLine, Inc. All rights reserved.