Speaker: Dr. Cynthia Kong

Title: Manufacturing of Luck - Examining Biased Inference of Luck from Field to the Lab

Luck is defined as "success or failure apparently brought by chance rather than through one's own actions." People tend to infer luck from past (random) events even if they are proven independent and often form biased beliefs. In the lottery markets, for example, past winning numbers are believed to be less lucky and thus receive a smaller share of the betting proportion in future draws, although winning numbers are drawn independently. This is the famous "gambler's fallacy" in lottery play often documented in previous works. Surprisingly, we demonstrate that the opposite effect, the hot-hand fallacy, can exist in lottery games, i.e., past winning numbers can be considered luckier and receive more bets. We first use data from two different lottery games to show that game design can alter players' inference of luck and that the hot-hand fallacy can exist in lottery games with multiple winning numbers. Second, we use a different lottery game to demonstrate significant differences in behavior between" preferred" and" non-preferred" options. In particular, while the gambler's fallacy persists among most non-preferred options, it disappears for most preferred options and even turns into hot-hand behavior. We test those hypotheses established from online data by online experimental studies and confirm that luck can be manufactured by altering the game design or introducing preferred labels on options. Our findings contribute to the existing literature on the influence of game designs on the occurrence of cognitive biases and build a bridge to research on preferences and the role of emotions in human decision-making.