It’s a famous thought experiment, popular at a certain kind of dinner party: “The Trolley Problem.” Let’s say you were given the job of operating the lever to a pair of train tracks on which a mine trolley is hurtling at breakneck speed. If you do nothing, the trolley will kill five people standing on the track it’s currently on. If you pull the lever, you’ll divert the trolley, and instead kill one person who is standing on the other track. The classic thinking holds that most people will let five people die rather than pull the lever, because the thought of deliberately killing one person — even to save five others — is intolerable to anyone with an ounce of empathy.
In a recent study, a group of researchers put this theory to the test, albeit with mouse subjects and imaginary scenarios instead of actual humans. The 300 subjects were first given a series of hypothetical moral dilemmas, including a variation on the trolley problem in which they were asked whether they’d push a man through a locked window to his certain death in order to make an escape for five children trapped inside. (One wonders, a little, about the people that think these things up.)
Later, two-thirds of the participants were taken to a lab, where they were shown two cages: one holding a single mouse, and another holding five. Subjects were told that if they did nothing, in 20 seconds the cage holding five mice would deliver a painful (but nonlethal) shock, but that if they pressed a button, they could divert the shock to the lone mouse in the other cage. (You may be happy to hear that no mice were to be shocked at all, but the participants did not know this at the time.) The other third of the subjects were given only a description of the experiment, and asked to give their hypothetical responses.
Interestingly, the subjects that were given the real-life experiment were more than twice as likely to press the button than those who were given the hypothetical option (34 percent as opposed to 16 percent, respectively). Most people still chose to do nothing rather than take a role in inflicting pain, but the findings suggest that what people say they’d do in a situation can be wildly different from what they do when they’re actually in it — in fact, the researchers found that the subjects’ early responses to the moral dilemma questionnaire couldn’t predict their eventual actions in the real-life mouse task. Nor did the subjects’ responses to a questionnaire on their empathy toward animals affect how they later behaved toward the mice.
The researchers speculate that these results suggest a human propensity toward virtue signaling, or expressing values that they believe to be correct, whether or not they actually hold those values in practice — a theory which is, uh, depressingly familiar in the current political climate.