Washing away good and bad luck: People believe it works

Zwick, working with Alison Jing Xu of the University of Toronto, and Norbert Schwarz of the University of Michigan, designed two experiments that showed risk taking depends on whether participants recalled a past episode of good or bad luck and whether they washed their hands before engaging in a risky decision making task.

The experimental findings, in the paper "Washing Away Your (Good or Bad) Luck: Physical Cleansing Affects Risk-Taking Behavior," converge with anecdotal reports of superstitious practices, such as an athlete wearing the same unwashed shirt during a winning streak, and show that magical beliefs about luck have behavioral consequences.

Magical beliefs are exhibited, for example, by having confidence in one's ability to predict the outcome of a random event beyond the known probabilities if one can exert irrelevant control on the situation. For example, research has shown people are more confident they will have a winning scratch-off lottery ticket if they pick the ticket instead of being given one by a clerk.

Debriefing conversations with participants suggest that people remain unaware of these influences, as has also been observed in other studies. Although participants are familiar with the underlying metaphors and related superstitious practices, they do not realize that this knowledge is applicable to the experiment and, needless to say, insist that they would never be influenced by such a thing.

In the first experiment outlined in the paper, 59 business students at a North American university were brought in. Half were asked to recall an incident in which they had good luck financially and the other half were asked to recall an incident of bad financial luck.

Next, an ostensibly unrelated product evaluation study served as a cover story. All participants evaluated an antiseptic wipe. Half were told to use the wipe, the other half were told to form and impression of the product without using it. Product evaluation questions followed.

Finally, the participants were given a managerial decision task. Taking the role of a chief executive officer, they had to adopt or reject a product improvement recommendation based on two consequences of action.

Under the first option, if they stayed with the existing product profits would remain at the current level, about $20 million per year.

Under the second option, the product was modified, but profits would depend on acceptance by consumers. Marketing research indicated there was a 75 percent chance of strong acceptance, which would result in an increase in profits to $24 million, but there was a 25 percent chance of weak acceptance, resulting in a drop in profits to $12 million.

The researchers found those who recalled an unlucky incident and cleaned their hands and those that recalled a lucky incident and didn't clean their hands were more likely to select the riskier option.

Of those who recalled an unlucky incident and cleaned their hands, 73 percent selected the riskier option, while only 36 percent who recalled an unlucky incident and didn't clean their hands picked the riskier option.

Of those who recalled a lucky incident, 77 percent who didn't clean their hands picked the riskier option, while only 35 percent who cleaned their hands selected the riskier option.

In the second experiment, students and staff from the Hong Kong University of Science and Technology, where Zwick formerly taught, were given HK $100 (US$1 = HK$7.8) to gamble with. They were told this was "for real" money that they would keep at the end. Indeed, they were paid based on their decisions and luck.

The experimenters showed participants a pink ball and a green ball and placed them in a bag. Participants selected one of the colors as their "winning" color and blindly picked a ball from the bag. If they picked the winning color they won HK$50. If not, they lost HK$50. They repeated the task until they lost their HK$100, won an additional HK$100 or completed four rounds.

Next, an ostensibly unrelated product evaluation study served as a cover story for the hand-washing manipulation. Participants evaluated organic soap. Half were told to wash their hands with the soap. The other half were told not to use the soap.

Finally, participants did a second round of gambling. They received HK$50 and were told they could bet any amount from nothing to HK$50.It was the same game as last time, but with only one round.

Researchers found participants who had good luck in the initial round bet more money in the second round than participants who had bad luck.

However, participants who had bad luck in the first round bet more money in the second round if they washed their hands. The difference was an average of HK$31.15 versus HK$17.47.

In contrast, those who had good luck in the first round bet less money in the second round if they had washed their hands. The difference was an average of HK$28.08 versus HK$37.75.