Amazon Mechanical Turk and Moral Psychology

Do check out this post over at Social Science++, titled Moral psychology on Amazon Mechanical Turk. Basically, the Amazon Mechanical Turk is a web service that lets you create simple online tasks or questionnaires and pay online workers to do them. The blogger, Brendan, creates a few moral dilemma scenarios, and asks people to say what they would do in that situation. The situations are all essentially "would you cause the death of one person in order to save five?". However, they vary from flicking a switch in order to redirect a deadly item from an area of high population to an area of low population (the trolley problem), to actually killing a young healthy man in order to harvest his organs to save the lives of others. Despite the rational equality of all the scenarios, they all get different responses. This suggests to us that moral reasoning is not rational, but emotional. Despite the title of "moral psychology", there are several philosophical issues at stake here.

1. There is obviously a concept of action here, that allowing something to happen is not the same as making something happen. (This does not hold for all moral codes, and many (including Asimov's laws of robotics) believe that causing an outcome through inaction is equivalent to causing an outcome through action.) But to what extent are we responsible for those actions we do not interfere with?
2. Is there a concept of "natural" or "inevitable" death at stake here, that makes the death of the organ needers that much more acceptable?
3. Do the split-second decision making required of the first two scenarios, and the speedy nature with which their consequences will unfold, have an effect on the judgement of the decisions? Is the third scenario, the organ-harvesting one, that much less acceptable because of the length of time invested in carrying out this "lesser of two evils"?

Unfortunately, moral philosophy isn't something I know a great deal about (although I hope to learn), so I can't say if there are any philosophical theories that speculate along these lines. If anyone does know, I'd be very interested to find out, so please, leave a comment.

On another note, the actual method of data-gathering as outlined in the original blog post is a very interesting one. I think greater demographic survey of the web community is needed before it can be fully validated as a method of collecting empirical data, but the project itself, especially the flexible scale of tasks, make it ideal for small squibs and breakfast experiments.