Alas, there is nothing new under the sun. I’m guessing the herding hypothesis also says that only very strong private signals can override the public ones too. So, if this is an old hypothesis well-known to you, why would you then lament the herding? If herding is the case, then not updating (much) after a certain point gives you better results than continuing to update, doesn’t it? And if it does, then wouldn’t that ‘win’ and be the rational thing to do given the circumstances?
(Alternate question: if not-updating is rational, why resort to social signalling explanations for the not-updating? Social signalling may explain how the herding starts and perpetuates itself, but there’s no need to drag it in as an explanation for not-updating as well.)
Science works hugely better than random crackpottery, but is also very far from optimal.
If you can’t trust yourself to update on evidence, then go with science. If you can (you’re here, aren’t you?) then updating will leave you better off.
You can always limit yourself to updating in all but the most obvious cases that science misses, and doing marginally better.
You can always limit yourself to updating in all but the most obvious cases that science misses, and doing marginally better.
No doubt that this is what many scientists do - ‘this is what I really think, but I’ll admit it’s not generally accepted’. But I’d put the emphasis on updating only in the obvious cases and otherwise trusting in science, because how many areas of science can one really know well enough to do better than the subject-area consensus?
This is the standard “herding” hypothesis, that public behavior ignore private signals once public signals have become lopsided enough.
Alas, there is nothing new under the sun. I’m guessing the herding hypothesis also says that only very strong private signals can override the public ones too. So, if this is an old hypothesis well-known to you, why would you then lament the herding? If herding is the case, then not updating (much) after a certain point gives you better results than continuing to update, doesn’t it? And if it does, then wouldn’t that ‘win’ and be the rational thing to do given the circumstances?
(Alternate question: if not-updating is rational, why resort to social signalling explanations for the not-updating? Social signalling may explain how the herding starts and perpetuates itself, but there’s no need to drag it in as an explanation for not-updating as well.)
This looks like the “Science vs Bayes” distinction to me.
Science works hugely better than random crackpottery, but is also very far from optimal.
If you can’t trust yourself to update on evidence, then go with science. If you can (you’re here, aren’t you?) then updating will leave you better off.
You can always limit yourself to updating in all but the most obvious cases that science misses, and doing marginally better.
No doubt that this is what many scientists do - ‘this is what I really think, but I’ll admit it’s not generally accepted’. But I’d put the emphasis on updating only in the obvious cases and otherwise trusting in science, because how many areas of science can one really know well enough to do better than the subject-area consensus?