Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
During the past 40 years there has been an accumulation of experimental evidence suggesting that most of the axioms of expected utility theory are liable to be systematically violated by substantial numbers of individuals. Much of this evidence has focused in failures of the independence axioms and has stimulated a number of alternative models that try to explain that evidence in various ways. This article presents a fresh experiments that looks at a different kind of violation--one that does not appear to be easily accommodated by several of the more prominent alternative models as they are currently formulated. Copyright 1991 by Kluwer Academic Publishers