(Edit) You can pretty much ignore this comment, but you should read pangloss’ responses below. I found them enlightening.
I am not a psychologist, but I do know that our current plan (of, for example, thinking about the brainteaser cases), is definitely not the way to develop actual expertise.
Personally, I think there are two extremes at studying something like this. One is to work from the edge cases toward the general/common cases and the other is to work from the general/common cases toward the edges. Some gradients in the middle certainly apply.
My hunch is that our perceptions of the common cases have been painted over by too many other ways of life/thinking. Rationality can have trouble getting its point through the layers. The edge cases make more sense to flesh out because the edge cases are explicitly designed to force specific ways of life/thinking into choosing between one or more well-defined options.
Edge cases also have the advantage of Jargon. We can make up our own words to mean something that we were not able to express previously and, by definition, be correct. While this is less amusing than describing it in laymen’s terms it certainly gets the problem defined. Working from the edges with new descriptions is a long process and one easy enough to mess up. Pigeonholing is a great risk and we can describing large swaths of general/common cases with a new word is dangerous.
General cases have the advantage that they can provide a foundation that will instantaneously apply to related cases. The network can build faster but the definitions may not be as clear-cut. A word used to describe wine can be used to describe faces (e.g. bitter) and people will be able to follow the conversation. The danger this time is generalizing terms and cases so that we are no longer able to distinguish between specific concepts.
All that being said, language is nifty. Learning how to work with it correctly would certainly be useful.
(Note) I am making a fine distinction between pigeonholing and generalizing that may exist only in my head. I see pigeonholing as taking large numbers of varied things and trying to separate them into ill-defined categories that force more variance than is really necessary. I see generalizing as taking large numbers of varied things and slopping them into categories that begin to erase the variances. If someone else knows better terms for this, please let me know.
I think the question about which cases to focus on when forming theories is different from the question of which cases to use to train oneself to verbalize one’s thoughts without interfering with one’s thinking. The latter requires us to train on paradigms, the former may be something we can pursue in either direction.
This is crucial: The thought isn’t to presuppose which direction our theorizing should go, but rather to make sure that when we theorize, we aren’t tripping ourselves up.
Mmm, very good point. Strangely, now that I think about it, this sound very similar to the concept of the highest principle:
You may try to name the highest principle with names such as “the map that reflects the territory” or “experience of success and failure” or “Bayesian decision theory”. But perhaps you describe incorrectly the nameless virtue. How will you discover your mistake? Not by comparing your description to itself, but by comparing it to that which you did not name.
In the comparison between Rationality Recognition and Face Recognition, what is the Rationality Recognition equivalent of sight?
It depends. Sometimes it will be sight or our other senses, sometimes it will be memory, sometimes it will be testimony.
Thinks about it this way, we take in information all the time, and draw conclusions from it. “Sight” isn’t playing a key role in face recognition except providing the data, you have a mental program for matching visual face data to previous visual face data, and that program gets screwed up if you start thinking through a description of the face after you see it.
Similarly, you see a room full of objects and events. You’ve got one or more “draw conclusions” programs that run on the data you see, and that program can get screwed up by putting things into words that you don’t normally.
The data on insight puzzles shows that if you do manage to draw the right conclusions, and you try to put into words how you did it, you may get screwed up in the following way: you are confident in explanation A for how you drew the conclusion, when, in actuality, the truth is radically different explanation B.
My claim isn’t about rationality recognition per se, it is simply this: psychology has shown that verbalizing can screw us up when dealing with a process that isn’t normally done verbally. And a lot (if not most) of our inferential processes are not done in this explicitly verbalized manner (verbalized doesn’t necessarily mean spoken aloud, but just ‘thinking through in words’).
My claim is that there are known ways to get good at verbalizing non-verbal processes, and they involve training on paradigmatic cases. It is only after such training that one can start thinking about edge cases and the borderlands without worrying that the process of discussing the cases is corrupting their thinking about the cases.
Before we can advance rationality by discussion, we must first learn to discuss rationality.
My claim isn’t about rationality recognition per se, it is simply this: psychology has shown that verbalizing can screw us up when dealing with a process that isn’t normally done verbally. And a lot (if not most) of our inferential processes are not done in this explicitly verbalized manner (verbalized doesn’t necessarily mean spoken aloud, but just ‘thinking through in words’).
My claim is that there are known ways to get good at verbalizing non-verbal processes, and they involve training on paradigmatic cases. It is only after such training that one can start thinking about edge cases and the borderlands without worrying that the process of discussing the cases is corrupting their thinking about the cases.
Before we can advance rationality by discussion, we must first learn to discuss rationality.
Understood. Thanks for the clarification. Going back and rereading the article after these comments made a few more lights click on in my head.
(Edit) You can pretty much ignore this comment, but you should read pangloss’ responses below. I found them enlightening.
Personally, I think there are two extremes at studying something like this. One is to work from the edge cases toward the general/common cases and the other is to work from the general/common cases toward the edges. Some gradients in the middle certainly apply.
My hunch is that our perceptions of the common cases have been painted over by too many other ways of life/thinking. Rationality can have trouble getting its point through the layers. The edge cases make more sense to flesh out because the edge cases are explicitly designed to force specific ways of life/thinking into choosing between one or more well-defined options.
Edge cases also have the advantage of Jargon. We can make up our own words to mean something that we were not able to express previously and, by definition, be correct. While this is less amusing than describing it in laymen’s terms it certainly gets the problem defined. Working from the edges with new descriptions is a long process and one easy enough to mess up. Pigeonholing is a great risk and we can describing large swaths of general/common cases with a new word is dangerous.
General cases have the advantage that they can provide a foundation that will instantaneously apply to related cases. The network can build faster but the definitions may not be as clear-cut. A word used to describe wine can be used to describe faces (e.g. bitter) and people will be able to follow the conversation. The danger this time is generalizing terms and cases so that we are no longer able to distinguish between specific concepts.
All that being said, language is nifty. Learning how to work with it correctly would certainly be useful.
(Note) I am making a fine distinction between pigeonholing and generalizing that may exist only in my head. I see pigeonholing as taking large numbers of varied things and trying to separate them into ill-defined categories that force more variance than is really necessary. I see generalizing as taking large numbers of varied things and slopping them into categories that begin to erase the variances. If someone else knows better terms for this, please let me know.
I think the question about which cases to focus on when forming theories is different from the question of which cases to use to train oneself to verbalize one’s thoughts without interfering with one’s thinking. The latter requires us to train on paradigms, the former may be something we can pursue in either direction.
This is crucial: The thought isn’t to presuppose which direction our theorizing should go, but rather to make sure that when we theorize, we aren’t tripping ourselves up.
Mmm, very good point. Strangely, now that I think about it, this sound very similar to the concept of the highest principle:
In the comparison between Rationality Recognition and Face Recognition, what is the Rationality Recognition equivalent of sight?
It depends. Sometimes it will be sight or our other senses, sometimes it will be memory, sometimes it will be testimony.
Thinks about it this way, we take in information all the time, and draw conclusions from it. “Sight” isn’t playing a key role in face recognition except providing the data, you have a mental program for matching visual face data to previous visual face data, and that program gets screwed up if you start thinking through a description of the face after you see it.
Similarly, you see a room full of objects and events. You’ve got one or more “draw conclusions” programs that run on the data you see, and that program can get screwed up by putting things into words that you don’t normally.
The data on insight puzzles shows that if you do manage to draw the right conclusions, and you try to put into words how you did it, you may get screwed up in the following way: you are confident in explanation A for how you drew the conclusion, when, in actuality, the truth is radically different explanation B.
My claim isn’t about rationality recognition per se, it is simply this: psychology has shown that verbalizing can screw us up when dealing with a process that isn’t normally done verbally. And a lot (if not most) of our inferential processes are not done in this explicitly verbalized manner (verbalized doesn’t necessarily mean spoken aloud, but just ‘thinking through in words’).
My claim is that there are known ways to get good at verbalizing non-verbal processes, and they involve training on paradigmatic cases. It is only after such training that one can start thinking about edge cases and the borderlands without worrying that the process of discussing the cases is corrupting their thinking about the cases.
Before we can advance rationality by discussion, we must first learn to discuss rationality.
Understood. Thanks for the clarification. Going back and rereading the article after these comments made a few more lights click on in my head.
So, where do we start?
I guess we find out how to acquire verbal expertise in a given domain, and do so for rationality, reasoning, and inference.