In article <CI9owz.Aw3@news.Hawaii.Edu> I write:
>In article <mike.756246065@motion> mike@psych.ualberta.ca
> (Mike Dawson) writes:
>> ...
>>Instead of following this up by posting your discovery (which would
>>have interesting consquences; for instance, I'd be interested in the
>>differences between how these two different groups came to believe in
>>X), ... [where X is ... whatever]
>
>The answer to your question about how NLPers come to believe certain
>things is that their ideas are based on insight gained from clinical
>experience. And one of the ways these ideas are tested is by doing
>exercises during NLP trainings.

Although this is largely true, it gives the erroneous impression that all the ideas and methods in NLP come merely from insight and intuition. In fact, something that all the NLP trainers I've experienced agree on (including such people as Richard Bandler, Leslie Cameron Bandler, David Gordon, Robert Dilts, and Steve & Connirae Andreas who have played key roles in developing the field) is that what NLP is really about is ``modeling,'' in fact what NLP really *is* is modeling, and that all the therapeutic ideas and techniques that NLP is well known for are actually by-products of modeling.

Modeling is the process of asking a person questions in order to understand their mental processes and in particular how they perform a particular mental skill. Learning this completely enough that one can teach others to perform the same skill. NLP has developed a lot of tools for asking these questions effectively, but I think I can summarize them all in two rules: go very very slowly; and know the right things to look for.

Going very very slowly is quite difficult to learn. I remember last summer during an exercise I volunteered to be modeled for my skill at handling crisis calls when I used to do suicide prevention work. As I started to go through a demonstration of handling a call, I said ``Of course most calls that we got were not crisis calls at all. A lot of them were lonely callers who just want to chat, and many of them were simply requesting information about something or asking for help with everyday problems. But when I realize that a call is a crisis call...'' At this point, Todd Epstein, who had wandered over to our group, interrupted to ask ``How do you know a call is a crisis call?'' And I said, ``Oh most of the time that's fairly evident right away. That was usually no problem.''

Part of the art of modeling is being alert to those moments where the subject says ``That's no problem'' or ``I just know'' or ``I just do it.'' The key to good modeling is being able to get information about precisely those things that the subject does so well that he doesn't even think about them and doesn't know how to explain them. In the example above, for instance, we wound up spending the whole exercise (about half an hour) studying how I was able to very quickly decide what kind of caller I was dealing with and have a high degree of accuracy in this judgement. And we only scratched the surface.

Another key point in modeling is that you can't learn much by listening to a subject talk about what he does in the abstract. You need to watch the subject actually do it --- or at least pretend to do it. So when my fellow students modeled me, they had me pretend to pick up the phone to deal with a call. (``Don't assume the caller is already on the line. Start when the phone rings.'') And the subject has to be willing to put up with the fact that you are going to constantly be interrupting him with questions.

The NLP spelling strategy was developed by asking a number of good spellers and poor spellers to spell some words, and then studying what they did differently from each other. In this case, as in many cases, eye accessing cues (``lateral eye movements,'' as apparently you guys prefer to say) are very helpful. It's not so much a matter of putting an interpretation on the eye movements (``Up and to the subject's right means he's constructing an image, up and to the left means he's recalling a stored image''). It's that the moment when the eyes move lets you know that it's time to interrupt with a question. (``What happened just now? What did you just do? Are you seeing a picture of something?'')

Other non-verbal cues can also be important. (``What does that mean --- that gesture you just made? What was going on in your mind when you made that gesture?'') (``I noticed that when you first started your leg kept jerking up and down fairly rapidly. And then at a certain point the leg stopped moving. Go back to the point when the leg stopped moving and tell me what was going on in your mind.'')

A lot of the most important stuff is below the threshhold of consciousness, so you really have to slow the subject down so he becomes aware of what he's doing.

According to what Cameron Bandler told us, the Visual-Kinesthetic Dissociation technique (the previous incarnation of the Fast Phobia/Trauma Cure) was developed because during a seminar in Chicago the NLP people were talking about bad memories and a woman in the audience said ``But I don't have any bad memories. None of my memories seem bad to me.'' And the NLPers thought that was fascinating and immediately started modeling her to find out how she could do that.

It is this art of asking a person questions about their subjective experience that I think would be above all of interest to cognitive psychologists. And I have to concede that it's the most difficult of all aspects of NLP to learn. The average person who ``does NLP'' has no competance at all in modeling.

But for the person who wants to scientifically study the structure of subjective experience, modeling is where the real juice is.