Picture the following hypothetical scenario: A trolley is headed toward five helpless victims. The trolley can be redirected so that only one person's life is at stake. Psychologists and philosophers have been using moral dilemmas like this for years asking, would you redirect the train? Is it morally acceptable to do this? Experts usually switch up the details to see how different sub-scenarios affect moral judgment. Many researchers have come to the conclusion that an individual's moral judgment in this type of scenario is strongly guided by abstract moral principles.
However, researchers of a study to appear in an upcoming issue of Cognitive Science see a problem with this approach, "Small changes in wording can affect judgments in ways that have nothing to do with differences in moral principles. Psychologists that analyze judgment and decision making in consumer behavior are aware of this fact. We applied these same methods to this scenario to illustrate that subjects' responses could not possibly be attributed to any known moral principles."
The study shows that it is often hard to distinguish between the influences of moral principles and more general biases. The authors make the claim that the conclusions of other researchers, which point to a strong moral influence, may be directly influenced by the researchers' disciplinary background (philosophy, etc.), rather than the morality of the individual.
When presented with moral dilemmas in which participants could sacrifice some people in order to save more, participants were highly sensitive to the proportion of lives saved (e.g. 8 out of 10 vs. 8 out of 40), and they demanded that more lives be saved when more lives were at risk, even though the number to be sacrificed remained constant. Participants are also less likely to support taking a moral action if they were asked to on their own generate more reasons for doing so.
The authors hope their research will lead to a more skeptical attitude towards drawing inferences regarding moral principles from studies of moral dilemmas, and that researchers will move to design other methods for studying moral judgments that move beyond the use of moral dilemmas.