Yes, they were quite poorly formed, which renders the study entirely pointless. The idea behind these problems is to collect people's intuitive reactions to si

Yes, they were quite poorly formed, which renders the study entirely pointless. The idea behind these problems is to collect people's intuitive reactions to simple, clear situations. If the problems are so poorly formed that intuition simply rejects them, the exercise is useless, except as a study in human frustration. (Perhaps that's the point? It would explain the poor UI design.) Several situations present you with the option of causing one immediate and fairly certain death to prevent several more distant, much more speculative deaths. You can't predict the path of a boulder down a mountain. A boulder that would be stopped by one person would be unlikely to kill five. The man driving the injured people to the hospital actually has little idea how quickly he needs to get them there to save their lives. The certainty presented in the problems is so unreal that they might as well have been posed in an entirely abstract way in the first place.