4 ms·
The first option makes sense to me. Both choices have the same expected value; I understand that, but I also feel it's more likely I have misunderstood somethin
by furyofantares 3y ago
The first option makes sense to me. Both choices have the same expected value; I understand that, but I also feel it's more likely I have misunderstood something about the second option. I'm crystal clear on the first.
Literally any tiebreaker for "which of these options with all the same value" seems fine, I do have to make a decision after all. And I think I'd be slightly skeptical that the experimenter is trying to trick me in some way with the second option in the unlikely case that I've misunderstood it.
- 6gvONxR4sf7o 3y agoI agree this seems to be the reason. The second urn has a more uncertain return in the same way that I'm uncertain about the 98th digit of pi. It's not a random event, but my certainty around it is related to the deductive effort I put in. It's immediately clear that they're the same-ish, and also immediately clear that the first urn doesn't have a gotcha. Still very interesting.
- p_j_w 3y agoThe experimenter can't trick you in the second option because you get to choose the color. Both options have the same expected value.
- furyofantares 3y agoI was clear that I understand that both options have the same expected value. I said so twice in the post.
- inimino 3y agoThis assumes the experimenter has zero ability to predict your choice.
- cubefox 3y agoBut presumably people would pick the first urn even if it had a bit worse expected utility than the second. Say, the green ball wins, and you can choose between either drawing from an urn with 29 green and 31 red balls, or drawing from an urn with an unknown number of red and green balls. The Wikipedia article says that people have active ambiguity aversion, not just that they use ambiguity as a tie breaker.