This seems to be extracting a lot of inference from not a lot of evidence. The difference in confidence could simply arise from the fact that the "2 dollars mor

This seems to be extracting a lot of inference from not a lot of evidence. The difference in confidence could simply arise from the fact that the "2 dollars more" question sounds harder than the control question. It's a more complicated question, so it stands to reason that people would be less confident about it. What I really want to know is this: how confident were the people who got the harder question right?