If evolution were optimizing creatures that were exactly like humans except that they were not social at all (maybe they each have their own planet), then almost all of the biases would disappear.
by Stephanie Zolayvar Dec 15 2016
If evolution were optimizing creatures that were exactly like humans except that they were not social at all (maybe they each have their own planet), then almost all of the biases would disappear.
Comments
Stephanie Zolayvar
Paul's recent post argues in favor of this position: 6mt.html
Alexei Andreev
I disagree mostly on priors, since it's quite unlikely that we discovered, understood, and pinpointed the biggest source. Would agree with a slightly weaker version of this claim.
Eric Bruylant
Does "better" include "more like the in-group"? If yes, this seems very plausible. If no, I'd guess the crony beliefs cluster is a bigger source.
Better in a locally zero-sum way has some direct checks (because the people you're interacting with you have an incentive to see if you're deceiving them about your usefulness), whereas locally positive-sum biases (e.g. "my in group is the best in-group, and is right about everything") should be selected for.