Peter

This is interesting, as your answers were among the more "Mitchelly" of the answers given - I think I would have guessed that you were one of the people less biased by the initial example, more willing to update your rule to take account of new information rather than to stay consistent with the first rule you thought of.

Feynman - if you're doing science, then arguably by definition you get to test your hypotheses, and the smart-aleck hypotheses have got to be worth testing.