I'd be curious to see how many studies controlled for relative emotional value of money, eg. by looking for differences in behavior for all groups at various values.

One could hypothesize that people might be more willing to share an unexpected "windfall", the perception of which could feasibly scale with one's emotional perception of wealth.