Last night I decided to experiment with omegle's spy mode. In this mode you pose a question and watch two anonymous strangers 'debate' it. The first thing to notice is the extremely poor quality of the 'questions' -- most of them are bald statements or requests for 'sex'. I decided I'd run a question to test 'dysrationalia' among the participants. My method was to use the question on cosmicvariance (with the names changed to make it less searchable):
Jack is looking at Anne, but Anne is looking at George. Jack is married, but George is not. Is a married person looking at an unmarried person?
A) Yes B) No C) Can't tell
If you haven't seen this before you should take a moment to try and work it out, I'll state the answer and the explanation a little further down to give you a chance to not accidentally see it.
I tested this question with probably about 100+ people, so I'll accept it is not a statistically significant result. I also had to remove all the people who simply disconnected or responded with a random answer and disconnected. I only counted people who said *something* in support of their argument. The overall result was that exactly 1 person got the correct answer. This correct answer, which was reasoned out on the spot rather than having been heard before, was the high point of the experience.
Now that you've had some space to work out the answer I can explain it. The correct answer is A. The reason is that Anne is either married or she is not. If she is not then Jack is looking at her (so true) and if she is then she is looking at George (so true). Statistically most of you will have said C, giving the reason that we do not know what state Anne is in. I think the reason why we stop there is because in most situations it is too expensive to enumerate all the possible states so we make the judgement that since we don't know Anne's state we don't know. This is an example of "dysrationalia". The concept is of course explained far better elsewhere and hopefully you now want to find out more.
Aside from the fact that the vast majority of people on omegle gave C as the answer, they also justified it in a very one-sided way. The explanation was given, roughly split equally, between "We don't know about Anne so C" and "Anne could be married so C". The latter answer flirts with the correct answer since they only need to actually think about what happens if Anne is married. That people were making statements and just not following through the logic was disappointing, but it is interesting that no-one said "Anne could be unmarried so C". This is rather indicative of some thought process that goes on when people think about this type of problem. There's actually another thing in logic where peoples thoughts on it have a similar (incorrect) asymmetry, but I can't remember it now (and haven't been able to for the last half hour...), but no doubt I will come to me *just* after I post this. The one person who gave the right answer initially said C, because of the second reason above, but then they followed the thought through and got the right answer.
The most depressing thing about the exercise was watching two people explain to each other how obvious it was that the answer was C. Eventually I had to quit for this reason, if you give someone a problem you should at least be able to help explain the solution so that they don't ingrain the wrong answer.