Experiments are a dime-a-dozen. Here is one: Get a bunch of groups of undergraduate students (try to choose within the same major because of the problem of self-selection into majors by value systems and personality traits). In both groups, each student will be given a short document containing spelling errors. Each spelling error found is worth, say, $2 (or maybe 2 points on the final but that is harder to get past Institutional Review Boards).
With some groups, announce that every person will earn the same amount equal to the $2 times the total number of spelling errors identified by the group divided by the number of people in the group.
With other groups, announce that everyone will be paid based on the number of spelling errors she finds and nothing else.
With other groups, announce that everyone will be paid a certain base amount (say, $4) plus ($2 - d) times per spelling error she finds.
With yet more groups, ask them to vote on the distribution rule. There are at least two treatments here: For example, you can ask the students to vote on the distribution rule before doing the work; if you do this, you can have treatments based on whether the results of the vote are announced. Or you can ask the students to vote on the distribution rule after doing the work.
Students must be paid in real money. You must not use any deception. (These are the rules that set Economics experiments apart from those Psychology experiments).
Try to guess what kind of relationship between "total output" and distribution rule.
Of course, there is intrinsic motivation. It may be tickled by a desire to make more money (so that one can afford better things for oneself or for others that one cares about -- a desire to make money does not preclude caring for others) or it may be tickled by other things the person values that cannot easily be bought with money.
I am reminded of my favorite line in the Aviator: "You do not care about money," Howard Hughes says to his fiancee's family, "because you have it."
Jeff (and Joel): I love SO. I have put a lot of time into answering questions on SO because I like it (see http://stackoverflow.com/users/100754/sinan-unur ). I have also mentioned on meta.stackoverflow.com that we are generate content so you and Joel can make money. I don't mind it. I get something out of SO which you cannot buy with money but yet is still valuable. Plus, I have been enjoying both of your blogs for almost a decade now.
Just stop pretending having money, wanting money etc are bad things. Enjoy yours.