The study tested how people make decisions about risky choices and prices. They found that when pricing gambles, people's risk attitudes matched a specific pattern. However, when choosing between a gamble and its expected value, their decisions seemed random and often opposite to the pattern. This was true for both individual people and across different subjects.