This week’s post is a comprehensive summary of the rules and criterion of the Good Judgment Project, adapted in service of evaluating one’s predictions.
commoncog.com/blog/how-do-yo
Conversation
One of the more interesting things, diving into The Good Judgment Project, is that the entire research program is really a fertile ground for testing related things like “How effective is cognitive debiasing? Can debiasing even work?”
1
1
1
We know that Kahneman doesn’t believe that overcoming cognitive biases is possible. What and co have discovered is that yes, it is possible, but primarily in the context of teams.
You can’t use training to debias. And you can’t really debias as a single individual.
1
Surprisingly, however, training serves to decrease noise, instead of helping with bias. knowledge.insead.edu/strategy/the-s
I can’t find an actual paper about this, probably because I haven’t searched hard enough. But this seems to be a very recent discovery.
Anyway, as this is a recent development, many of the things Tetlock, Mellers and co have said about overcoming cognitive biases needs rethinking. They’re changing their minds about the mechanism — we should too.
(Also note that Kahneman’s been working on a book … about noise!)