The homeopath who has treated thousands of patients, should listen to the high-school chemistry student who has evidence that homeopathy doesn’t work. The physics crackpot who has worked on their theory of everything for decades should listen to the student of physics who points out that it fails to predict the results of an experiment.
Wow.
After reading just what was presented in the anecdote, you have strong enough belief that the submitter was wrong about her own mind, and her programmer boyfriend was right, that you’ll compare her to frauds and crackpots whose ideas have vanishingly small probability.
you have strong enough belief that the submitter was wrong about her own mind,
and her programmer boyfriend was right
No, no, certainly not, I made it clear that I was arguing in general and could not comment on the specific example given (come on, I say this twice in the post you quote).
that you’ll compare her to frauds and crackpots whose ideas have vanishingly small probability.
Where do you get that probability mass from?
Let me repeat the argument she made
I am the one who has spent millions of minutes in this mind, able to directly experience what’s going on inside of it.
This sort of argument, “I have observed this phenomenon for far longer than you did, therefore I am vastly more likely to be right about this than you are”, is very vulnerable to confirmation bias (among other biases), where the speaker will more easily remember events that fit her hypothesis than events which didn’t. This argument is a stereotypical crackpot argument, I gave two examples but I can (alas) give many more. It is virtually never a good argument. Someone who is actually sitting on top of mountains of evidence for a certain hypothesis need not resort to this argument, they can just show the evidence!
How often have I seen crackpots use this argument? Dozens of times. How often have I seen non-crackpots use it? I recall only one occasion, two if you include the OP. How often have I seen people who have actually carefully collected lots of evidence use this argument? Never. (Is my memory on this subject susceptible to confirmation bias? Ha! Yes, of course it is.). Is it any wonder then, that my prior for “people who use this argument are crackpots” is somewhat large?
How is this relevant to the example given? We cannot expect everyone to continuously gather relatively unbiased evidence on their own behaviour, can we? Indeed we cannot. Then, we should also not be extremely confident in the models of ourselves which we have constructed. If someone challenges these models, what should we do?
Most likely, the person challenging our models does not actually have good evidence and is just attempting to make some status move. This is the most common and least interesting possibility, ignoring him / breaking up with him / telling him to stop doing it …. may all be good courses of action (yes, I disagree less with the OP than you may think)
If evidence is actually put forward (which it wasn’t in the OP example, but which I hope it would be on less wrong), you can provide evidence of your own “but in the past, when X happened, I did Y, which is compatible with my self-model but not with your model of me”. Ideally, the arguers should update after the exchange of evidence. (“I observed myself for millions of minutes” does not count as evidence exchange, since the other person already knew that)
I was not arguing about the specific example given in the OP
After reading just what was presented in the anecdote, you have strong enough belief that the submitter was wrong about her own mind, and her programmer boyfriend was right, that you’ll compare her to frauds and crackpots whose ideas have vanishingly small probability.
Wow.
After reading just what was presented in the anecdote, you have strong enough belief that the submitter was wrong about her own mind, and her programmer boyfriend was right, that you’ll compare her to frauds and crackpots whose ideas have vanishingly small probability.
Where do you get that probability mass from?
No, no, certainly not, I made it clear that I was arguing in general and could not comment on the specific example given (come on, I say this twice in the post you quote).
Let me repeat the argument she made
This sort of argument, “I have observed this phenomenon for far longer than you did, therefore I am vastly more likely to be right about this than you are”, is very vulnerable to confirmation bias (among other biases), where the speaker will more easily remember events that fit her hypothesis than events which didn’t. This argument is a stereotypical crackpot argument, I gave two examples but I can (alas) give many more. It is virtually never a good argument. Someone who is actually sitting on top of mountains of evidence for a certain hypothesis need not resort to this argument, they can just show the evidence!
How often have I seen crackpots use this argument? Dozens of times. How often have I seen non-crackpots use it? I recall only one occasion, two if you include the OP. How often have I seen people who have actually carefully collected lots of evidence use this argument? Never. (Is my memory on this subject susceptible to confirmation bias? Ha! Yes, of course it is.). Is it any wonder then, that my prior for “people who use this argument are crackpots” is somewhat large?
How is this relevant to the example given? We cannot expect everyone to continuously gather relatively unbiased evidence on their own behaviour, can we? Indeed we cannot. Then, we should also not be extremely confident in the models of ourselves which we have constructed. If someone challenges these models, what should we do?
Most likely, the person challenging our models does not actually have good evidence and is just attempting to make some status move. This is the most common and least interesting possibility, ignoring him / breaking up with him / telling him to stop doing it …. may all be good courses of action (yes, I disagree less with the OP than you may think)
If evidence is actually put forward (which it wasn’t in the OP example, but which I hope it would be on less wrong), you can provide evidence of your own “but in the past, when X happened, I did Y, which is compatible with my self-model but not with your model of me”. Ideally, the arguers should update after the exchange of evidence. (“I observed myself for millions of minutes” does not count as evidence exchange, since the other person already knew that)
...