It seems to me that industrial organization and industrial psychology have put quite a bit of effort into asking how to get committees and groups to think together effectively. Perhaps someone could do a literature survey / find some good books to review for LW?
If Mercier and Sperber’s theory is correct, people are already optimized for arguing things out in groups ..which would mean that rationality training is really solo rationality training...and perhaps not that useful for many people.
If Mercier and Sperber’s theory is correct, people are already optimized for arguing things out in groups
Not really, no. People are optimized for winning arguments against untrained humans. The point of group rationality training is figuring out what norms / individual training / etc. makes it so that the best ideas (by some external metric) are most likely to win in a group discussion, rather than the best-championed ideas. Even if, say, I can identify why someone’s argument is not helping push towards truth, there needs to be a group norm that I can call them out on that and that will be effective. (Think of “Objection!” or pointing out fallacies in debate club; both of those rest on the common acceptance on what things are worth objecting to or calling fallacious.)
The average person isn’t as well optimized at group debate that the best debates, but people are still optimized for group debate in the sense of individual pondering.
It seems to me that industrial organization and industrial psychology have put quite a bit of effort into asking how to get committees and groups to think together effectively. Perhaps someone could do a literature survey / find some good books to review for LW?
If Mercier and Sperber’s theory is correct, people are already optimized for arguing things out in groups ..which would mean that rationality training is really solo rationality training...and perhaps not that useful for many people.
Not really, no. People are optimized for winning arguments against untrained humans. The point of group rationality training is figuring out what norms / individual training / etc. makes it so that the best ideas (by some external metric) are most likely to win in a group discussion, rather than the best-championed ideas. Even if, say, I can identify why someone’s argument is not helping push towards truth, there needs to be a group norm that I can call them out on that and that will be effective. (Think of “Objection!” or pointing out fallacies in debate club; both of those rest on the common acceptance on what things are worth objecting to or calling fallacious.)
The average person isn’t as well optimized at group debate that the best debates, but people are still optimized for group debate in the sense of individual pondering.