4. Swish. "Now 'swish' these two pictures. Start with seeing that cue picture, big and bright. Then put a small, dark image of the outcome picture in the lower right corner. The small dark image will grow big and bright and cover the first picture, which will get dim and shrink away as fast as you can say 'swish.' Then blank out the screen, or open your eyes. Swish it again a total of five times. Be sure to blank the screen at the end of each swish."

5. Test.

a. "Now picture that first image. . . . What happens?" If the swish has been effective, this will be hard to do. The picture will tend to fade away and be replaced by the second image of yourself as you want to be.

b. Another way to test is behavioral: Find a way to create the cues that are represented in your partner's cue picture. If that picture is of your partner's own behavior, as it was with Jack, ask him to actually do it. If that picture is of someone else offering a chocolate or a cigarette, or yelling, then I want you to do that with your partner and observe what she does and how she responds.

If the old behavior is still there when you test, back up and do the swish pattern again. See if you can figure out what you left out, or what else you can do to make this process work. I'm teaching you a very simple version of a much more general pattern. I know that some of you have questions, but I want you to try doing this first before asking them. After you've actually tried it, your questions will be much more interesting. Take about fifteen minutes each. Go ahead.

As I went around the room, I observed many of you succeeding. Let's not talk about that unless you had difficulty and then came up with something interesting that made it work. I want to hear about when it didn't work.

Amy: I want to stop smoking. But when we tested, I still have the urge to smoke.

OK. Describe your first picture for me.

Amy: I see myself with a cigarette in my mouth, and—

Stop. It's very important that you don't see yourself in that first picture, and that you do see yourself in the second picture. That's an essential part of what makes the swish work. The first picture has to be an associated image of what you see out of your own eyes as you start to smoke — your hand reaching for a cigarette, for instance. If you see your hand with a cigarette in it, do you feel compulsed to smoke? Or is it seeing the cigarettes? Whatever it is, I want you to make a picture of what you see that fires off that feeling of wanting to smoke; make a picture of whatever precedes smoking. It might be reaching for a cigarette, lighting it, bringing it up to your lips, or whatever else you do. Try the process with that picture, and report back.

Man: Which book has this process in it?

None. Why would I teach you something that is already in a book? You're adults; you can read, I have always thought it was really idiotic for someone to write a book and then to go and read it to people at seminars. But a lot of people do exactly that, and some of them make a lot of money, so I guess it has some use.

Woman: In a lot of the earlier NLP techniques you substitute a specific new behavior. But in this one you just see the way you'd be different if you changed.

That's right. That's what makes this pattern so generative. Rather than substituting a specific behavior, you're creating a direction. You're using what's often called "self–image," a very powerful motivator, to set that direction.

When I was in Toronto in January, a woman said she had a phobia of worms. Since Toronto is frozen over most of the year, I didn't think that would be too much of a problem, so I said, "Well, why don't you just avoid them?" She said "Well, it just doesn't fit with the way I see myself." That mismatch motivated her very strongly, even though the worm phobia wasn't actually that much of a problem for her. It wasn't even what I call a "flaming phobia." It was an "ahhh!" phobia, rather than an "AAHHGGH!" phobia. She didn't yet have her brain directionalized appropriately, but that image of herself kept her trying. So of course I asked her "If you made this change, how would you see yourself differently?" The effectiveness of this pattern depends most importantly on getting the answer to that question. This process doesn't get you to an endpoint — it propels you in a particular direction. If you saw yourself doing something in particular, you would only program in that one new choice. If you see yourself being a person with different qualities, that new person can generate many new specific possibilities. Once you set that direction, the person will start generating specific behavior faster than you can believe.

If I had used the standard phobia cure on her, she wouldn't care about worms at all, and wouldn't even notice them. To get somebody to not care about something is too easy, and there is enough of that going on in the world already. If I had built in a specific behavior, like picking up worms, then she'd be able to pick up worms. Neither of those changes are particularly profound in terms of this woman's personal evolution. It seems to me that there are more interesting changes that a human being can make.

When I swished her, I set up a direction so that she is drawn toward that image of herself as more competent, happier, more capable, liking herself better, and most important, able to believe that she can quickly make changes in the way she wants to.

Woman: I think I understand that, but I'm trying to fit it in with some of the NLP anchoring techniques I learned earlier. For instance, there's a technique where you make a picture of how you'd like to be, then step into it to get the kinesthetic feelings, and then anchor that state.

Right. That's one of the old techniques. It has its uses, but it also has certain drawbacks. If someone has a really detailed and accurate internal representation, you can create a specific behavior that will work very nicely. But if you just make a picture of what you'd like to be like, and step inside it to feel what it's like to be there, that doesn't necessarily mean that you are there with any quality, or that you learned much along the way. It's an excellent way to build self–delusions, and it also doesn't give you anywhere else to go.

A lot of people go to therapists asking to feel more confident, when they're incompetent. That lack of confidence may be accurate feedback about their abilities. If you use anchoring to make someone feel confident, that feeling may allow her to do things she actually could do already but wasn't confident enough to try. That will increase her abilities as well. But it may only create overconfidence — someone who is still incompetent but doesn't notice it any more! There are plenty of people like that around already, and they're often dangerous to others as well as to themselves. I've been commenting for years about how many people ask a therapist for confidence, and so few ask for competence.

You can change somebody so that he believes he's the very best at something he does, when he can't do it very well at all. When a person is good at acting confident, he usually convinces lots of other people to trust in abilities that he doesn't have. It never ceases to amaze me how many people think that if an "expert" acts confident, he must know what he is doing. I figure that as long as you are going to have a false sense of security, you might as well develop some competence along the way.

Where's Amy? Did you finish doing the swish with the new picture?

Amy: Yes.

How long did it take you to do that five times? Amy: Quite a while.