One way to frame this would be getting really good at learning tacit knowledge.
Is there some reason to believe that being good at “simulating the felt senses of their conversational partners in their own minds” (whatever this means—still unclear to me) leads to being “really good at learning tacit knowledge”? In fact, is there any reason to believe that being “really good at learning tacit knowledge” is a thing?
One way would be to interact one with them, notice “hey, this person is really good at this” and then inquire as to how they got so good. This is my experience with seasoned authentic relaters.
Hmm, so in your experience, “seasoned authentic relaters” are really good at “simulating the felt senses of their conversational partners in their own minds”—is that right? If so, then the followup question is: is there some way for me to come into possession of evidence of this claim’s truth, without personally interacting with many (or any) “seasoned authentic relaters”?
Another way would be to realize there’s a hole in understanding related to intuitions
Can you say more about how you came to realize this?
Let’s say that as a designer, you wanted to impart your intuition of what makes good design. Would you rather have:
A newbie designer who has spent hundreds of hours of deliberate practice understanding and being able to transfer models of how someone is feeling/relating to different concepts, and being able to model them in their own mind.
A newbie designer who hasn’t done that.
Well, my first step would be to stop wanting that, because it is not a sensible (or, perhaps, even coherent) thing to want.
However, supposing that I nevertheless persisted in wanting to “impart my intuition”, I would definitely rather have #2 than #1. I would expect that having done what you describe in #1 would hinder, rather than help, the accomplishment of this sort of goal.
Is there some reason to believe that being good at “simulating the felt senses of their conversational partners in their own minds” (whatever this means—still unclear to me) leads to being “really good at learning tacit knowledge”?
This requires some model of how intuitions work. One model I like to use is to think about “intuition” is like a felt sense or aesthetic that relates to hundreds of little associations you’re picking up from a particular situation.
If i’m quickly able to in my mind, get a sense for what it feels like for you (i.e get that same felt sense or aesthetic feel when looking at what you’re looking at), and use circling like tools to be able to tease out which parts of the environment most contribute to that aesthetic feel, I can quickly create similar associations in my own mind and thus develop similar intuitions.
f so, then the followup question is: is there some way for me to come into possession of evidence of this claim’s truth, without personally interacting with many (or any) “seasoned authentic relaters”?
Possibly you could update by hearing many other people who have interacted with seasoned authentic relaters stating they believe this to be the case.
Can you say more about how you came to realize this?
I mean, to me this was just obvious seeing for instance how little emphasis the rationalists I interact with emphasize things like deliberate practice relative to things like conversation and explicit thinking. I’m not sure how CFAR recognized it.
However, supposing that I nevertheless persisted in wanting to “impart my intuition”, I would definitely rather have #2 than #1. I would expect that having done what you describe in #1 would hinder, rather than help, the accomplishment of this sort of goal.
I think this is a coherent stance if you think the general “learning intuitions” skill is impossible. But imagine if it weren’t, would you agree that training it would be useful?
This requires some model of how intuitions work. One model I like to use is […]
Hmm. It’s possible that I don’t understand what you mean by “felt sense”. Do you have a link to any discussion of this term / concept?
That aside, the model you have sketched seems implausible to me; but, more to the point, I wonder what rent it pays? Perhaps it might predict, for example, that certain people might be really good at learning tacit knowledge, etc.; but then the obvious question becomes: fair enough, and how do we test these predictions?
In other words, “my model of intuitions predicts X” is not a sufficient reason to believe X, unless those predictions have been borne out somehow, or the model validated empirically, or both. As always, some examples would be useful.
Possibly you could update by hearing many other people who have interacted with seasoned authentic relaters stating they believe this to be the case.
It is not clear to me whether this would be evidence (in the strict Bayesian sense); is it more likely that the people from whom I have heard such things would make these claims if they were true than otherwise? I am genuinely unsure, but even if the answer is yes, the odds ratio is low; if evidence, it’s a very weak form thereof.
Conversely, if this sort of thing is the only form of evidence put forth, then that itself is evidence against, as it were!
I mean, to me this was just obvious seeing for instance how little emphasis the rationalists I interact with emphasize things like deliberate practice relative to things like conversation and explicit thinking. I’m not sure how CFAR recognized it.
Hmm, I am inclined to agree with your observation re: deliberate practice. It does seem puzzling to me that the solution to the (reasonable) view “intuition is undervalued, and as a consequence deliberate practice is under-emphasized” would be “let’s try to understand intuition, via circling etc.” rather than “let’s develop intuitions, via deliberate practice, whereupon the results will speak for themselves, and this will also lead to improved understanding”. (Corollary question: have the efforts made toward understanding intuitions yielded an improved emphasis on deliberate practice, and have the results thereof been positive and obvious?)
I think this is a coherent stance if you think the general “learning intuitions” skill is impossible. But imagine if it weren’t, would you agree that training it would be useful?
Indeed, I would, but notice that what you’re asking is different than what you asked before.
In your earlier comment, you asked whether I would find it useful (in the hypothetical “newbie designer” situation) to be dealing with someone who had undertaken a lot of “deliberate practice understanding and being able to transfer models of how someone is feeling/relating to different concepts, and being able to model them in their own mind”.
Now, you are asking whether I judge “training … the general ‘learning intuitions’ skill” to be useful.
Your questions imply that these are the same thing. But (even in the hypothetical case where there is such a thing as the latter) they are not!
The wikipedia article for Gendlin’s focusing has a section trying to describe felt sense, taking out the specific part about “the body”, the first part says:
” Gendlin gave the name “felt sense” to the unclear, pre-verbal sense of “something”—the inner knowledge or awareness that has not been consciously thought or verbalized”,
Which is fairly close to my use of it here.
That aside, the model you have sketched seems implausible to me; but, more to the point, I wonder what rent it pays? Perhaps it might predict, for example, that certain people might be really good at learning tacit knowledge, etc.; but then the obvious question becomes: fair enough, and how do we test these predictions?
One thing it might predict is that there are ways to train the transfer of intuition, from both the teaching and learning side of things, and that by teaching them people get better at picking up intuitions.
Hmm, I am inclined to agree with your observation re: deliberate practice. It does seem puzzling to me that the solution to the (reasonable) view “intuition is undervalued, and as a consequence deliberate practice is under-emphasized”
I do believe CFAR at one point was teaching deliberate practice and calling it “turbocharged training”. However, if one is really interested in intiution and thinks its’ useful, the next obvious step is to ask “ok, I have this blunt instrument for teaching intuition called deliberate practice, can we use an understanding of how intuitions work to improve upon it?”
Your questions imply that these are the same thing. But (even in the hypothetical case where there is such a thing as the latter) they are not!
Good catch, this assumes that my simplified model of how intuitions work is at least partly correct. If the felt sense you get from a particular situation doesn’t relate to intuition, or if its’ impossible for one human being to get better at feeling what another is feeling, than these are not equivalent. I happen to think both are true.
One thing it might predict is that there are ways to train the transfer of intuition, from both the teaching and learning side of things, and that by teaching them people get better at picking up intuitions.
Well, my question stands. That is a prediction, sure (if a vague one), but now how do we test it? What concrete observations would we expect, and which are excluded, etc.? What has actually been observed? I’m talking specifics, now; data or case studies—but in any case very concrete evidence, not generalities!
I do believe CFAR at one point was teaching deliberate practice and calling it “turbocharged training”. However, if one is really interested in intiution and thinks its’ useful, the next obvious step is to ask “ok, I have this blunt instrument for teaching intuition called deliberate practice, can we use an understanding of how intuitions work to improve upon it?”
Yes… perhaps this is true. Yet in this case, we would expect to continue to use the available instruments (however blunt they may be) until such time as sharper tools are (a) available, and (b) have been firmly established as being more effective than the blunt ones. But it seems to me like neither (a) (if I’m reading your “at one point” comment correctly), nor (b), is the case here?
Really, what I don’t think I’ve seen, in this discussion, is any of what I, in a previous comment, referred to as “the cake”. This continues to trouble me!
I suspect the CFARians have more delicious cake for you, as I haven’t put that much time into circling, and the related connection skills I worked on more than a decade ago and have atrophied since.
Things I remember:
much quicker connection with people
there was a few things like exercise that I wasn’t passionate about but wanted to be. After talking with people who were passionate I was able to become passionate myself for those things
I was able to more quickly learn social cognitive strategies by interacting with others who had them.
To suggest something more concrete… would you predict that if an X-ist wanted to pass a Y-ist’s ITT, they would have more success if the two of them sat down to circle beforehand? Relative to doing nothing, and/or relative to other possible interventions like discussing X vs Y? For values of X and Y like Democrat/Republican, yay-SJ/boo-SJ, cat person/dog person, MIRI’s approach to AI/Paul Christiano’s approach?
It seems to me that (roughly speaking) if circling was more successful than other interventions, or successful on a wider range of topics, that would validate its utility. Said, do you agree?
I always think of ‘felt sense’ as, not just pre-verbal intuitions, but intuitions associated with physical sensations, be they in my head, shoulders, stomach, etc.
Yeah, same; I think this term has experienced some semantic drift, which is confusing. I meant to refer to pre-verbal intuitions in general, not just ones accompanied by physical sensation.
Yes, I think there’s a distinction between the semantic content of “My intuition is that Design A is better than Design B” referring to the semantic content or how the intuition “caches out” in terms of decisions. This contrast with the felt sense, which always seems to refer to what the intuition is like “from the inside,” for example a sense of unease when looking at Design A, and rightness when looking at Design B.
I feel like using the word “intuition” can refer to both the latter and the former, whereas when I say “felt sense” it always refers to the latter.
Is there some reason to believe that being good at “simulating the felt senses of their conversational partners in their own minds” (whatever this means—still unclear to me) leads to being “really good at learning tacit knowledge”? In fact, is there any reason to believe that being “really good at learning tacit knowledge” is a thing?
Hmm, so in your experience, “seasoned authentic relaters” are really good at “simulating the felt senses of their conversational partners in their own minds”—is that right? If so, then the followup question is: is there some way for me to come into possession of evidence of this claim’s truth, without personally interacting with many (or any) “seasoned authentic relaters”?
Can you say more about how you came to realize this?
Well, my first step would be to stop wanting that, because it is not a sensible (or, perhaps, even coherent) thing to want.
However, supposing that I nevertheless persisted in wanting to “impart my intuition”, I would definitely rather have #2 than #1. I would expect that having done what you describe in #1 would hinder, rather than help, the accomplishment of this sort of goal.
This requires some model of how intuitions work. One model I like to use is to think about “intuition” is like a felt sense or aesthetic that relates to hundreds of little associations you’re picking up from a particular situation.
If i’m quickly able to in my mind, get a sense for what it feels like for you (i.e get that same felt sense or aesthetic feel when looking at what you’re looking at), and use circling like tools to be able to tease out which parts of the environment most contribute to that aesthetic feel, I can quickly create similar associations in my own mind and thus develop similar intuitions.
Possibly you could update by hearing many other people who have interacted with seasoned authentic relaters stating they believe this to be the case.
I mean, to me this was just obvious seeing for instance how little emphasis the rationalists I interact with emphasize things like deliberate practice relative to things like conversation and explicit thinking. I’m not sure how CFAR recognized it.
I think this is a coherent stance if you think the general “learning intuitions” skill is impossible. But imagine if it weren’t, would you agree that training it would be useful?
Hmm. It’s possible that I don’t understand what you mean by “felt sense”. Do you have a link to any discussion of this term / concept?
That aside, the model you have sketched seems implausible to me; but, more to the point, I wonder what rent it pays? Perhaps it might predict, for example, that certain people might be really good at learning tacit knowledge, etc.; but then the obvious question becomes: fair enough, and how do we test these predictions?
In other words, “my model of intuitions predicts X” is not a sufficient reason to believe X, unless those predictions have been borne out somehow, or the model validated empirically, or both. As always, some examples would be useful.
It is not clear to me whether this would be evidence (in the strict Bayesian sense); is it more likely that the people from whom I have heard such things would make these claims if they were true than otherwise? I am genuinely unsure, but even if the answer is yes, the odds ratio is low; if evidence, it’s a very weak form thereof.
Conversely, if this sort of thing is the only form of evidence put forth, then that itself is evidence against, as it were!
Hmm, I am inclined to agree with your observation re: deliberate practice. It does seem puzzling to me that the solution to the (reasonable) view “intuition is undervalued, and as a consequence deliberate practice is under-emphasized” would be “let’s try to understand intuition, via circling etc.” rather than “let’s develop intuitions, via deliberate practice, whereupon the results will speak for themselves, and this will also lead to improved understanding”. (Corollary question: have the efforts made toward understanding intuitions yielded an improved emphasis on deliberate practice, and have the results thereof been positive and obvious?)
Indeed, I would, but notice that what you’re asking is different than what you asked before.
In your earlier comment, you asked whether I would find it useful (in the hypothetical “newbie designer” situation) to be dealing with someone who had undertaken a lot of “deliberate practice understanding and being able to transfer models of how someone is feeling/relating to different concepts, and being able to model them in their own mind”.
Now, you are asking whether I judge “training … the general ‘learning intuitions’ skill” to be useful.
Your questions imply that these are the same thing. But (even in the hypothetical case where there is such a thing as the latter) they are not!
The wikipedia article for Gendlin’s focusing has a section trying to describe felt sense, taking out the specific part about “the body”, the first part says:
” Gendlin gave the name “felt sense” to the unclear, pre-verbal sense of “something”—the inner knowledge or awareness that has not been consciously thought or verbalized”,
Which is fairly close to my use of it here.
One thing it might predict is that there are ways to train the transfer of intuition, from both the teaching and learning side of things, and that by teaching them people get better at picking up intuitions.
I do believe CFAR at one point was teaching deliberate practice and calling it “turbocharged training”. However, if one is really interested in intiution and thinks its’ useful, the next obvious step is to ask “ok, I have this blunt instrument for teaching intuition called deliberate practice, can we use an understanding of how intuitions work to improve upon it?”
Good catch, this assumes that my simplified model of how intuitions work is at least partly correct. If the felt sense you get from a particular situation doesn’t relate to intuition, or if its’ impossible for one human being to get better at feeling what another is feeling, than these are not equivalent. I happen to think both are true.
I see, thanks.
Well, my question stands. That is a prediction, sure (if a vague one), but now how do we test it? What concrete observations would we expect, and which are excluded, etc.? What has actually been observed? I’m talking specifics, now; data or case studies—but in any case very concrete evidence, not generalities!
Yes… perhaps this is true. Yet in this case, we would expect to continue to use the available instruments (however blunt they may be) until such time as sharper tools are (a) available, and (b) have been firmly established as being more effective than the blunt ones. But it seems to me like neither (a) (if I’m reading your “at one point” comment correctly), nor (b), is the case here?
Really, what I don’t think I’ve seen, in this discussion, is any of what I, in a previous comment, referred to as “the cake”. This continues to trouble me!
I suspect the CFARians have more delicious cake for you, as I haven’t put that much time into circling, and the related connection skills I worked on more than a decade ago and have atrophied since.
Things I remember:
much quicker connection with people
there was a few things like exercise that I wasn’t passionate about but wanted to be. After talking with people who were passionate I was able to become passionate myself for those things
I was able to more quickly learn social cognitive strategies by interacting with others who had them.
To suggest something more concrete… would you predict that if an X-ist wanted to pass a Y-ist’s ITT, they would have more success if the two of them sat down to circle beforehand? Relative to doing nothing, and/or relative to other possible interventions like discussing X vs Y? For values of X and Y like Democrat/Republican, yay-SJ/boo-SJ, cat person/dog person, MIRI’s approach to AI/Paul Christiano’s approach?
It seems to me that (roughly speaking) if circling was more successful than other interventions, or successful on a wider range of topics, that would validate its utility. Said, do you agree?
Yes, although I expect the utility of circling over other methods to be dependent on the degree to which the ITT is based on intuitions.
I always think of ‘felt sense’ as, not just pre-verbal intuitions, but intuitions associated with physical sensations, be they in my head, shoulders, stomach, etc.
I think that Gendlin thinks all pre-verbal intuitions are represented with physical sensations.
I don’t agree with him but still use the felt-sense language in these parts because rationalists seem to know what I’m talking about.
Yeah, same; I think this term has experienced some semantic drift, which is confusing. I meant to refer to pre-verbal intuitions in general, not just ones accompanied by physical sensation.
Also in particular—felt sense refers to the qualia related to intuitions, rather than the intuitions themselves.
(Unsure, but I’m suspicious that the distinction between these two things might not be clear).
Yes, I think there’s a distinction between the semantic content of “My intuition is that Design A is better than Design B” referring to the semantic content or how the intuition “caches out” in terms of decisions. This contrast with the felt sense, which always seems to refer to what the intuition is like “from the inside,” for example a sense of unease when looking at Design A, and rightness when looking at Design B.
I feel like using the word “intuition” can refer to both the latter and the former, whereas when I say “felt sense” it always refers to the latter.