But conducting a controlled experiment and quantifying the result, instead of just going by anecdotal evidence about what worked for who, really is necessary.
Necessary for determining true theories, yes. Necessary for one individual to improve their own condition, no. If a mechanic uses the controlled experiment in place of his or her own observation and testing, that is a major fail.
“Just try my things!” you say,
I’ve been saying to try something. Anything. Just test something. Yes, I’ve suggested some ways for testing things, and some things to test. But most of them are not MY things, as I’ve said over and over and over.
At this point I’ve pretty much come to the conclusion that it’s impossible for me to discuss anything related to this topic on LW without this pervasive frame that I am trying to convince people to “try my things”… when in fact I’ve bent over backwards to point as much as possible to other people’s things. Believe it or not, I didn’t come here to promote my work or business.
I don’t care if you test my things. They’re not “my” things anyway. I’m annoyed that you think I don’t understand science, because it shows you’re rounding to the nearest cliche.
I actually advocate using a much higher standard of empirical testing of change techniques than is normally used in measuring psychological processes: observation of somatic markers (see Wikipedia re: the “somatic marker hypothesis”, if you haven’t previously).
Unlike self-reporting via questionnaire, many somatic markers can be treated as objective measures of results, because they are externally visible (facial expressions, posture change, etc.) and thus can be observed and measured by third parties. We can all agree whether someone flinches or grimaces or hangs their head in response to a statement—we are not dependent on the person themselves to tell us their internal reaction, nor do we have to sort through their conscious attempts to make their initial reaction look better.
True, I do not have a quantified scale for these markers, but it is nonetheless quantifiable—and it’s a direct outgrowth of a promising current neuroscience hypothesis. We can certainly observe whether a response is repeatable, and whether it is changed by any given intervention.
If someone wanted to turn that into controlled science, they’d have a lot of work ahead of them, but it could be done, and it would be a good idea. The catch, of course, is that you’d need to validate a somatic marker scale against some other, more subjective scale that’s already accepted, possibly in the context of some therapy that’s also relatively-validated. It seems to me that there are some chicken-and-egg problems there, but nothing that can’t be done in principle.
When I advocate that people try things, I mean that they should employ more-objective means of measurement—and on far-shorter timescales—than are traditionally used in the self-help field.
When I test some newfangled self-help modality (e.g. EFT, Sedona, etc.) it usually doesn’t take more than 30 minutes after learning the technique to know if it’s any good or not, because I have a way of measuring it that doesn’t depend on me doing any guessing. Either I still flinch or I don’t. Either I get a sinking feeling in my gut or I don’t. I know right then, in less time than it would take to list all the holes in their crazy pseudoscience theories about how the technique is supposed to work. (EFT, for example, works for certain things but its theory is on a par with Anton Mesmer’s theory of animal magnetism.)
I don’t know how you can get any more objective than that, at the level of individual testing. So, if there is anything that I’ve consistently advocated here, is that it’s possible to test self-help techniques by way of empirical observation of somatic marker responses both “before” and “after”. But even this is not “my” idea.
The somatic marker hypothesis is cutting-edge neuroscience—it still has a long way to go to reach the status of accepted theory. That makes using it as a testing method a bit more bleeding edge.
But for individual use, it has the advantage of being eminently testable.
Regarding the rest of your comment, I don’t see how I can respond, since as far as I can tell, you’re attacking things I never said… and if I had said them, I would agree with your impeccable critique of them. But since I didn’t say them… I don’t see what else I can possibly say.
Necessary for determining true theories, yes. Necessary for one individual to improve their own condition, no. If a mechanic uses the controlled experiment in place of his or her own observation and testing, that is a major fail.
I’ve been saying to try something. Anything. Just test something. Yes, I’ve suggested some ways for testing things, and some things to test. But most of them are not MY things, as I’ve said over and over and over.
At this point I’ve pretty much come to the conclusion that it’s impossible for me to discuss anything related to this topic on LW without this pervasive frame that I am trying to convince people to “try my things”… when in fact I’ve bent over backwards to point as much as possible to other people’s things. Believe it or not, I didn’t come here to promote my work or business.
I don’t care if you test my things. They’re not “my” things anyway. I’m annoyed that you think I don’t understand science, because it shows you’re rounding to the nearest cliche.
I actually advocate using a much higher standard of empirical testing of change techniques than is normally used in measuring psychological processes: observation of somatic markers (see Wikipedia re: the “somatic marker hypothesis”, if you haven’t previously).
Unlike self-reporting via questionnaire, many somatic markers can be treated as objective measures of results, because they are externally visible (facial expressions, posture change, etc.) and thus can be observed and measured by third parties. We can all agree whether someone flinches or grimaces or hangs their head in response to a statement—we are not dependent on the person themselves to tell us their internal reaction, nor do we have to sort through their conscious attempts to make their initial reaction look better.
True, I do not have a quantified scale for these markers, but it is nonetheless quantifiable—and it’s a direct outgrowth of a promising current neuroscience hypothesis. We can certainly observe whether a response is repeatable, and whether it is changed by any given intervention.
If someone wanted to turn that into controlled science, they’d have a lot of work ahead of them, but it could be done, and it would be a good idea. The catch, of course, is that you’d need to validate a somatic marker scale against some other, more subjective scale that’s already accepted, possibly in the context of some therapy that’s also relatively-validated. It seems to me that there are some chicken-and-egg problems there, but nothing that can’t be done in principle.
When I advocate that people try things, I mean that they should employ more-objective means of measurement—and on far-shorter timescales—than are traditionally used in the self-help field.
When I test some newfangled self-help modality (e.g. EFT, Sedona, etc.) it usually doesn’t take more than 30 minutes after learning the technique to know if it’s any good or not, because I have a way of measuring it that doesn’t depend on me doing any guessing. Either I still flinch or I don’t. Either I get a sinking feeling in my gut or I don’t. I know right then, in less time than it would take to list all the holes in their crazy pseudoscience theories about how the technique is supposed to work. (EFT, for example, works for certain things but its theory is on a par with Anton Mesmer’s theory of animal magnetism.)
I don’t know how you can get any more objective than that, at the level of individual testing. So, if there is anything that I’ve consistently advocated here, is that it’s possible to test self-help techniques by way of empirical observation of somatic marker responses both “before” and “after”. But even this is not “my” idea.
The somatic marker hypothesis is cutting-edge neuroscience—it still has a long way to go to reach the status of accepted theory. That makes using it as a testing method a bit more bleeding edge.
But for individual use, it has the advantage of being eminently testable.
Regarding the rest of your comment, I don’t see how I can respond, since as far as I can tell, you’re attacking things I never said… and if I had said them, I would agree with your impeccable critique of them. But since I didn’t say them… I don’t see what else I can possibly say.