I'm interested in knowing how you would estimate witness reliability who claims to have seen a miracle in the following case:
Linda says that she met with Santa Claus yesterday. She promises you that she's not lying.
Linda has financial incentives to claim that she met with Santa Claus yesterday. She will also not suffer any losses if she is found out to be lying. By using game theory we predict that the optimal strategy for Linda then will be to claim to have met Santa Claus yesterday even if it has not happened.
P(Lindas testimony | Linda did not meet with Santa Claus yesterday) = This one is unknown to us and we need to estimate it using common sense, logic, and philosophy.
My take is that since we don't know if Linda is giving us a false positive along with her financial incentives to lie, it would be a mistake to neglect the risk of a false positive, committing miss-rate neglect, and put a number below 0,5. What are your thoughts on this? Is there any perspective I'm missing out on?