Using Survey Evidence to Unleash System One Bias

There seem to be a couple of problems with using survey evidence. One problem is that because biases are intuitive or fast system responses, by definition people aren't conscious of them. The second problem is if we draw out the fallacy clearly enough to reveal the investors belief in it, then respondents could be educated out of the fallacy by the very questioning intended to uncover it. So how do you think about using survey evidence in light of some of these issues?