Purpose: While an individual's risk preferences are assumed to influence decisions that involve uncertainty, risk perceptions may also play a role. Belief in personal luck is one factor that may influence how individuals perceive risk. In this preliminary work, we examine the relationship between belief in luck and two kinds of decisions involving uncertainty, in a sample of cardiovascular patients.

Methods: In the first decision task, subjects make real choices over monetary lotteries and are paid their earnings. The second decision task involves hypothetical choices over two medications: one which will reduce the frequency of these symptoms with certainty, the other represents a gamble between complete relief of symptoms and no effect. Subjects also completed a survey measuring five concepts related to belief in luck: personal good luck, personal bad luck, general belief in luck, belief in the luck of others and the influence of luck on decision making.

Results: 79 patients who presented for stress testing were evaluated. We find that belief in luck varies across subjects; a significant number of subjects believe in luck. While 58% of subjects strongly or somewhat agreed with the statement “It is a mistake to base any decisions on how lucky you feel,” this is not consistent with the behavior we observed. We found significant positive correlations between belief in luck and the number of risky choices made in the lottery game and the health choices game. In the lottery game, there was a significant positive correlation between belief in others' luck and the number of times the subject's certainty equivalent was greater than the expected value of the lottery (ρ=.224, p<.05). Similarly with the health choices game, there was a significant positive correlation between both the number of times the subject chose the risky drug and belief in others' luck (ρ=.220, p=.052) and general belief in luck (ρ=.327, p<.01).

Conclusions: These results suggest that measurements of patients' risk preferences may also reflect patients' risk perceptions, which may in turn be influenced by belief in luck. By measuring patients' belief in luck, we enhanced our ability to distinguish between individuals who truly have risk seeking preferences and those who are either risk averse or risk neutral but whose perception of risk is mitigated by their belief in luck.