How does uncertainty influence moral behaviors?

We study whether individuals engage in more moral behaviors when faced with uncertain outcomes. We conduct two online experiments in which we manipulate the level of uncertainty associated with potential payoffs and observe participants’ choices in a subsequent moral decision. In the first experiment, the moral decision is the same as the decision which determines the uncertain payoff. In the second experiment, we relax the direct connection by separating the moral decision and the decision that determines the uncertainty payoff.

2024-04-29

2024-05-31

The primary outcome variable is a dummy which equals 1 if subject i choose the box with additional RMB 4 in choice problem c and 0 otherwise

This can be directly observed from the collected data. When the selected box is the same as the box with additional money in that round, the variable equals 1, otherwise it equals 0.

-Belief about winning the high bonus in the lottery when the person conducted moral or immoral behaviors
-Survey questions about the perception of the connection between moral behaviors and outcome of uncertainty.

We study whether individuals engage in more moral behaviors when faced with uncertain outcomes. We conduct two online experiments in which we manipulate the level of uncertainty associated with potential payoffs and observe participants’ choices in a subsequent moral decision. In the first experiment, the moral decision is the same as the decision which determines the uncertain payoff. In the second experiment, we relax the direct connection by separating the moral decision and the decision that determines the uncertainty payoff.

Experiment 1:
Part 1
In Part 1 of the experiment, subjects receive a lottery (h, n/6 ; l) in the form of six boxes numbered from 1 to 6 with n box(es) containing h(high payoff) and 6−n box(es) containing l(low payoff). They know the composition( how many boxes contain h and how many boxes contain l) but they do NOT know the exact distribution( which boxes contain h and which boxes contain l) in each round. There is uncertainty in payoff when n is not equal to 0 or 6. First, subjects are asked to choose a box and record the number they just chose on a paper(although they will never be asked to show the paper to experimenters, and they know about that). Then, subjects are informed of the presence of an additional 4 yuan(Chinese currency, referred to as RMB 4 hereafter) in one exact box out of the six. They know exactly which box contains the additional RMB 4. Then subjects are asked to report their initial box selection to receive the corresponding payoff in that box. Reporting the box containing the RMB 4(reporting +4) indicates either the truth based on their initial choice which happens to coincide with the box containing additional money (with a 1/6 chance), or a lie to maximize payoffs. Although lying cannot be observed individually, it can be measured at the aggregate level by the difference between the actual proportion of reporting +4 and 1/6.
There are 21 rounds in total, varying the distribution of bonus among the six boxes and which box contains additional money. There are three spreads between h and l: (40, n/6 ; 0), (30, n/6 ; 10), (22, n/6 ; 18), and seven levels of winning probability, p ∈ {0, 1/6 , 2/6 , 3/6 , 4/6, 5/6 , 1}.
The distribution of bonus among six boxes and which box contains additional money are given randomly and independently. Which box contains additional RMB 4 in each round was predetermined before the experiment starts with the RANDBETWEEN(1,6) function in Excel. The distribution of bonus among six boxes will be determined after the subjects finish all the decisions. One decision round will be randomly selected to be payoff relevant. Then the experimenter will use the RANDBETWEEN function in Excel to determine which boxes contain high payoff and the remaining boxes will contain low payoff. The experimenter will do the randomization of bonus according to the specific winning probability in the randomly selected round in front of the subjects after they complete the experiment. The above information is known to all the subjects before the start of the experiment.
Part 2
In Part 2 of the experiment, we elicit subjects’ belief about winning the high bonus in the lottery under different occasions. We ask the subjects to consider the decision scenario, where participants in group 1 lied in the experiment to receive additional money and participants in group 2 did not lie, reporting the previously recorded box truthfully and did not receive the additional money. We then ask the subjects which group if participants they believe is more likely to win the high bonus in the box lottery. If people behave more morally under uncer- tainty because they perceive a connection between their moral behaviors and the outcome of uncertainty, we should expect people to believe that the more honest group is more likely to win the high bonus. This question is not incentivized to avoid unnecessary complication. And since the question is about other subjects instead of the subjects themselves, we tend to believe that the subjects will answer truthfully even without incentives.
Part 3
In Part 3, we ask some survey questions about the perception of the connection between moral behaviors and outcome of uncertainty. We do not use any specific term such as karma or immanent justice reasoning in the survey. First, we describe the belief of the connection between uncertainty and morality and give an explicit example of the belief: Anna got extra change from the supermarket after shopping, although she was completely sure that no one would ever find it out, she decided to return the money when she remembered that she was waiting for the application result from her dream school. Then we ask whether the subjects know people who hold the belief(”Do you know anyone in your life who holds this kind of belief?”), and the proportion of people they believe to hold the belief(”What proportion of people do you think hold this kind of belief?”). Then we also ask if they hold the belief themselves(”Do you hold this kind of belief yourself?”), the extent to which the belief influence their daily actions(”To what extent does this kind of belief influence your actions and choices in daily life? ”), the extent to which they believe the belief is true(”To what extent do you think this kind of belief is true? That is, how likely is it that the world actually works according to this kind of belief.”). In addition, we ask whether they believe the world is a just place(”To what extent do you believe that the world is a just place (i.e. people get what they deserve)?”), and their general risk attitude(”How willing or unwilling are you to take risks in general?”).

Experiment 2:
Part 1
The general experimental manipulation is similar to the replication treatment, we manipulate the level of payoff uncertainty, and study people’s moral decision under different
7
levels of uncertainty. The difference is that subjects need to make two decisions separately: one decision determines their payoff(with or without uncertainty), and one moral decision.
First, subjects receive a lottery(h,n/6;l): with probability n/6 they will get the high payoff; with probability 1 − n/6 they will receive the low payoff. The lottery is in the form of choosing one brick from six bricks. There are different amount of payoff hiding behind different bricks. n bricks hide h(high payoff) and 6 − n bricks hide l(low payoff). Subjects know the composition( how many bricks hide h and how many bricks hide l) but they do NOT know the exact distribution(which bricks hide h and which bricks hide l) in each round. There is uncertainty in payoff when n ̸= 0, 6. Same as the replication treatment, there are 21 rounds in total, varying the distribution of bonus among the six bricks. There are three spreads between h and l: (40, n/6 ; 0), (30, n/6 ; 10), (22, n/6 ; 18), and seven levels of winning probability, p ∈ {0, 1/6 , 2/6 , 3/6 , 4/6, 5/6 , 1}. Subjects are asked to choose one brick in each round and submit their choice. They can not change their choice after the submission in each round. At the end of the experiment, one decision round will be randomly selected to be payoff relevant, and the subject will get the bonus behind the chosen brick in that round.
After the brick choice in each round (before the resolve of the uncertainty about their payoff), in a separate task, subjects are given another lottery in the form of choosing one box out of six boxes. They are informed that one exact box out of six contains RMB 4 in each round. The subjects are then asked to choose a box and get the bonus inside. Same as in replication treatment, they need to choose a box and record the number they just chose on a paper(although they will never be asked to show the paper to experimenters, and they know about that). After this step, on a new screen, they will be informed which exact box out of the six contains the RMB 4. Then subjects are asked to report their initial box selection to receive the corresponding payoff in that box. Again, reporting the box containing the RMB 4(reporting +4) indicates either the truth based on their initial choice which happens to coincide with the box containing the money (with a 1/6 chance), or a lie to maximize payoffs. Although lying cannot be observed individually, it can be measured at the aggregate level by the difference between the actual proportion of reporting +4 and 1/6.
The distribution of bonus among the six bricks and which box out of six contains RMB 4 are given randomly and independently. Which box contains RMB 4 in each round was predetermined before the experiment starts with the RANDBETWEEN(1,6) function in Excel. The distribution of bonus among six bricks will be determined after the subjects finish all the decisions. One decision round will be randomly selected to be payoff relevant. Then the experimenter will use the RANDBETWEEN function in Excel to determine which bricks hide high payoff and the remaining bricks will hide low payoff. The experimenter will do the randomization of bonus according to the specific winning probability in the randomly selected round in front of the subjects after they complete the experiment. The
above information is known to all the subjects before the start of the experiment.
Different from the replication treatment, this separation treatment does not suffer from regret aversion concern by design. Because the only bonus in the six boxes is the RMB 4(there is no h or l bonus distributed among the boxes), the subjects will have nothing to regret even if they change their initial choice of box. So whether to change their initial choice and report the box with additional money is purely a moral decision in this treatment. Besides, even if the regret confounder did not drive the results and uncertainty does motivate morality in the replication treatment, it is important to know whether people behave more morally under any type of uncertainty in general, or people only behave more morally under uncertainty that is directly connected to their moral decision. The separation treatment is a test of ”uncertainty motivates morality” under a more general setting, which represents
more prevalent situations in daily life.

Part 2 and Part 3 are the same as Experiment 1.

We coded the experiment with Qualtrics and will run the experiment with a Chinese experimental platform. The randomization will be done by computer.

Round level. Each subject goes through 21 rounds. The payoff uncertainty differs across the rounds. One round is randomly selected to be payoff-relevant.

Yes

240 subjects (21 rounds for each individual, cluster at individual level)

5040 observations (21 rounds per subject, 240*21 = 5040)

120 subjects per treatment, 240 subjects in total