Decision
The Allais paradox
Maurice Allais, 1953, in a paper arguing against the axioms of expected utility.
A lottery has 100 numbered tickets, one of which will be drawn. First choice. Lottery A pays 100 million with certainty. Lottery B pays 500 million if one of tickets 1 to 10 is drawn, 100 million if one of tickets 12 to 100 is drawn, and nothing if ticket 11 is drawn. Most people pick A. Second choice, same tickets. Lottery C pays 100 million if one of tickets 1 to 11 is drawn and nothing otherwise. Lottery D pays 500 million if one of tickets 1 to 10 is drawn and nothing otherwise. Most people pick D.
Look at tickets 12 to 100. In the first pair they pay 100 million under both A and B; in the second pair they pay nothing under both C and D. Everything else is identical across the two pairs. Can a person who picks A and D be choosing consistently?