90 of it is probably that they're thinking there must be a more involved solution than something a 7yr old could solve, otherwise no reasonable person would use

90 of it is probably that they're thinking there must be a more involved solution than something a 7yr old could solve, otherwise no reasonable person would use it to differentiate candidates with a university degree for a professional position. As an option, you could try prefacing the game with saying "the solution to this might be very easy", and see how many more solve it. Then you'd know you're likely not making a decision based on any signal of skill or ability at all, and just on them trying to make a prediction on what you're looking for. Edit: I had to come back to this, because the more I think about it the more it's mind boggling to me. You interview is very likely filtering out the smartest candidates and selecting for the rest. A typical strong candidate, that's prepped on interview questions, has maybe had other interviews at companies with challenging problems is going to look at it and think "This is a problem this interviewer uses to differentiate between top tier candidates and those who don't get offers. There is no way they're asking me a question that's so easy that any 7 year old child can answer it. There's gotta be something I'm missing." And keeps digging into it. And a very weak candidate will say think "This is my 10th interview, and sweet, finally a question I can answer!" These absurd backwards mindgames are exactly (one of the many things) thats wrong with software interviewing.