There's another experiment I like, this one done with trainee preists. Basically, you get four groups of preists and get them to write an essay. Half of them have to write an essay on 'The Good Samaritun' the other half on another bible story. Then you send them across campus to read their essay to some examiner/bigwig. You tell half the trainee preists to hurry as 'they're late' and you say nothing to the other one. Then, on the priest's path you put a researcher, 'passed out' pretending to be someone who was injured or unwell. Apparently, the main factor over whether the priests would stop and tend the man (despite the fact that preists in general are supposed to try and be good and help people) was whether they were late. Seeding their minds by making them thing about the parable of the good Samaritan didn't help. The late preists even stepped over the unconscious man, gripping their essay, with no thought for irony.

There's more experiments, such as the one where you can get ordinary people to act like sadists simply by making them prison guards or the one where people would torture another person just because they're told to by someone in authority.

So, really, you can design a lot of human behaviour by modifying the surroundings. And, the long distance train service (especially the one between winch and reading) is designed to cause conflict. They've shrunk the size of the train and filled it with reservations. I've noticed that as soon as people see that there aren't many seats, their behaviour changes from amenable and polite to people competing over scarce resources. I bet the 'pushy arrogant' people would have been more pleasant (although perhaps still arrogant) had they seen lots of seats.