You're right... I would demand some kind of evidence of success from a teacher. But if these prerequisites are at all easier to come by than the real thing, there's going to be a lot of faking going on.

Well, in the case of at least marketing and pickup, you can generally observe the teacher's own results, as long as you're being taught directly. For acting, you could observe the ability of the teacher's students. Copywriting teachers (people who teach the writing of direct marketing ads) can generally give sales statistics comparisons of their improvements over established "controls". (Btw, in the direct marketing industry, the "control" is just whatever ad you're currently using; it's not a control condition where you don't advertise or run a placebo ad!)

IOW, the practical arts of persuasion and belief do involve at least some empirical basis. One might quibble about what great or excellent acting or pickup might be, but anybody can tell bad acting or failed pickup. And marketing is measurable in dollars spent and actions taken. Marketers don't always understand math or how to use it, but they're motivated to use statistical tools for split-testing.

If your point was merely "some deluded people win"... then I'm not sure I get your point.

The ancient Greeks thought fire was an element, but that didn't stop them from using fire. Developing a practical model and a "true" theory are quite often independent things. My point is that you don't need a true theory to build useful models, or to learn and use them. And in most practical arts related to belief or persuasion, you will need to "act as if" certain beliefs are true, whether or not they are, because those beliefs nonetheless represent a model for reproducing behaviors that produce results under some set of circumstances.

For example, Seth Roberts' theory of calorie-flavor association is probably not entirely true -- but acting as if it were true produces results for some people under some circumstances. This represents progress, not failure.

"I first need a true belief T that U is useful".

Right -- and my process for that, with respect to self-help techniques, is mainly to look at the claims for a technique, and sort for ones that can be empirically verified and claim comparable or improved benefits relative to the ones that I've already tried. Assuming that the cost in time to learn the technique is reasonable (say, a few hours), and it can be implemented and tested quickly, that's sufficient T probability for me to engage in a test.

I can't help but notice that all of the arguments you just gave (that I pretty much agree with) for unquestioningly accepting a belief you've already decided to experimentally swallow... work equally well for theism. So what is it exactly, if not some flavor of T, that allows me to distinguish between the two?

Religion doesn't claim repeatable empirical benefits -- in fact they pretty carefully disclaim any. Zen is one of the few religions that contain procedures with claimed empirical benefits (e.g. meditation producing improved concentration and peace of mind), and those claims have actually held up pretty well under scientific investigation as well as my personal experimentation.

So, for me at least, your "T" consists mostly of claimed empirical benefits via a repeatable procedure capable of very short evaluation times -- preferably suitable for immediate evaluation of whether something worked or it didn't.

I do have two things that most people evaluating such things don't. At first, I tried a lot of these same techniques before I understood monoidealism and somatic markers, and couldn't get them to work. But once I had even the rudiments of those ideas -- not as theory but as experience -- I got many of the same things to work quite well.

That suggests very strongly to me that the major hidden variable in interpersonal variation of self-help technique applicability has less to do with the techniques themselves or any inherent property of the learner, than whether or not they've learned to distinguish conscious and unconcsious thoughts, and their abstract conception of an emotion or event from from its physical representation as a body sensation or as an internal image or sound. Most people (IME) seem to naturally confuse their internal narration about their experiences, and the experiences themselves. (Sort of like in "Drawing On The Right Side Of The Brain", where people confuse their symbols or abstractions for faces and hair with what they're actually seeing.)

Separating these things out are the primary skills I teach (as a vehicle to make other self-help techniques accessible) and many people require some sort of live feedback in order to learn them. There is some mild anecdotal evidence that prior experience with meditation helps -- i.e. the students who pick them up faster seem somewhat more likely to report prior meditation experience. But I haven't even tried to be rigorous about investigating that, since even non-meditators can learn the skill.

(Hm, now that I've written this, though, I wonder whether some of the Drawing On The Right Side Of The Brain exercises might be helpful in teaching these skills. I'll have to look into that.)

My, you are confident in your theories of human motivation.

If you look closely at what I said, I was explaining why I thought what I thought about your response, not saying that my thought was correct; I just wanted to explain why I had the impression that I did, not justify the impression or argue that it was actually true. That's a subtlety that's hard to convey in text, I suppose.