A real-world instance of Moore’s Paradox (“It’s raining, but I don’t believe it is”) occurs several times annually at Autzen Stadium in Eugene, Oregon —
Since 1990, Don Essig, the stadium’s PA announcer since 1968, has declared that “It never rains at Autzen Stadium” before each home game as the crowd chants along in unison. He often prefaces it with the local weather forecast, which quite often includes some chance of showers, but reminds fans that “we know the real forecast...” or “let’s tell our friends from (visiting team name) the real forecast...” If rain is actually falling before the game, Essig will often dismiss it as “a light drizzle”, or “liquid sunshine” but not actual rain by Oregon standards.
 Baker, Mark (March 6, 2010). “Still Quackin’”. The Register-Guard. Archived from the original on September 23, 2010. Retrieved September 20, 2010.
[link to archive:]
The Denes-Raj/Epstein study makes me wonder whether the subjects would still have picked the jar with 100 beans (7 red) if, say, the other jar had been announced to contain 6 beans (5 red) . Is there any “tipping point” (any specific number or percentage of red beans versus other beans) at which the subjects finally choose to follow the probabilities instead of going with “more reds”?
What if the other jar had been stated to contain only 5, 4, 3, 2, or 1 bean — but with ALL beans in that jar stated to be red? Would some subjects still go for the jar with 7 red beans in 100 (because 7 is more than five)? Has anyone tested the possibility that some subjects would actually say: “Yes, I know that I’m guaranteed to win if I pick from a jar that contain only one red bean and no other beans — but I’m still picking from the jar that has 7 red beans and 93 that aren’t red, because 7 is so much more than 1”?!