Hm, looking at the problems they used, I can actually see why people might fall for this. I've had to deal with a lot of people who seem reasonably intelligent overall but who let their biases get in the way of their reasoning.
For that matter, I could see myself falling prey to such assumptions, at least some of the time. I don't think I would have gotten the math wrong - except where I would have legitimately gotten it wrong
- but I would be more likely to double-check a result that I disagreed with than one I agreed with. That's almost exactly what they say in the study.
So, I guess the conclusion is that we should strive to check answers we agree with as fervently as we check answers we disagree with.
EDIT: In relation to the memory thing, I got into a somewhat heated argument with my roommate the day before yesterday regarding something that we remembered differently. We both agreed on the actual events, but disagreed on when they happened. As it happened, I was the one not remembering correctly (though I had to check with someone else to make sure). Which is not especially surprising, since it isn't the first time that I've gotten something like that mixed up in my head. It's been a stressful five months, too, which doesn't help.