Evidence sharing alone drives groups toward uniform belief without conformity psychology

Evidence

One of the most startling findings from the Bala-Goyal models is just how strongly people’s beliefs can influence one another. If we imagined a group of agents with no network connections gathering probabilistic evidence (and not sharing it), we would expect some of them to end up with the right theory and some with the wrong one. For instance, scientists who play the better slot machine and happen to lose all their money may give up on that machine for good. But with no communication, we should not expect much correlation between the various scientists’ beliefs. Some scientists would have good luck and stick with the better machine; others would not. Each conclusion would be independent from all the others. … Once scientists start to share evidence, however, it becomes extremely likely that they will all come to believe the same thing, for better or worse. Notice that this happens in the models only because the scientists share evidence. There is no psychology here. No one is imitating anyone else, no one is trying to conform, no one is smarter or dumber than the others. There are no thought leaders or sheeple. — Cailin O’Connor and James Owen Weatherall, The Misinformation Age, 208

Categorization

  • QUESTION: The claim that consensus emerges from pure rational evidence-sharing with no conformity psychology is striking — worth probing whether anything resembling a rational-agent account could explain real echo chambers as well as psychological accounts can.
  • XREF: connects to: Bayes’ rule is the unique rational way to update belief — Cailin O’Connor and James Owen Weatherall, The Misinformation Age: How False…, 108; Rational inference from others’ actions alone can push a whole group wrong — Cailin O’Connor and James Owen Weatherall, The Misinformation Age: How False…, 286; Consensual beliefs gain perceived validity and social connection from being shared (+4 more)

Broader Connections