results.
For present purposes, two general results are particularly germane:
1) It is possible to very power of a specific subject systematically and (within limits) arbitrarily in an experimental setting.
This can be done by manipulating some elements of his reputation or by manipulating some elements of his power experience.
This apparently innocuous-and certainly minimal-result is in fact not so unimportant.
It permits us to reject certain kinds of social choice models for certain kinds of situations.