Tracking down the seat of moral reasoning

Moral philosophers have long grappled with ethical questions, creating hypotheticals that test basic beliefs about right and wrong.  For example:

A trolley is running down a track out of control. If it keeps going, it will run over the five unsuspecting people hanging out on the track. You can prevent this disaster by throwing a switch, redirecting the trolley onto a siding where it will kill one person. Do you hit the switch? 

“Most people say that that’s OK, not great, but OK, and I have the same intuition that most people have,” says Joshua D. Greene, an Assistant Professor of Psychology in Harvard’s Faculty of Arts and Sciences. Greene is interested not only in what answers people give to these sorts of questions, but also in what kinds of intuitions drive such moral decision making and which regions of the brain they involve. And he is attempting to answer those questions by literally looking inside the brains of volunteers as they grapple with classic moral dilemmas.

One might think that the answers to Greene’s questions wouldn’t be affected by slight variations in the philosophical constructs; but one would be wrong. For instance, suppose the trolley problem is changed somewhat:

You’re watching the first drama unfold from a footbridge and the only way to save the five people from certain death is to push a large person standing next to you off of the bridge and onto the tracks. “He’ll get crushed by the trolley and die, but the five people will live, so it’s another five-over-one trade-off,” explains Greene, “but there it seems like it’s wrong.” The question is: Why do we think it’s all right to sacrifice one life for five in one case, but not in the other, and is there a good justification for the difference in our intuitions?

Expanding a new area of exploration

According to Dartmouth College philosophy professor Walter Sinnott-Armstrong, philosophers have traditionally been divided on the matter of how we come to have these intuitions. Some, like Immanuel Kant and his followers, believed moral judgments were rational, reasoned. Others like David Hume and his followers believed they were emotional judgments. But Harvard’s Greene picks up where classical philosophers left off, testing the hypothesis that both camps are right by asking people to make these judgments while recording their brain processes with a functional magnetic resonance imaging (fMRI) scanner. In so doing, says Dartmouth's Sinnott-Armstrong, Greene has made real a new field — neurophilosophy.

In Greene’s first neuroimaging study, volunteers responded to questions that appeared as text on a screen, which they viewed from inside the scanner using a mirror placed in front of their eyes. The machine recorded an image of the brain every one to three seconds. In this study, Greene saw increased activity in the medial prefrontal cortex, which is involved in emotion and social behavior, when volunteers considered cases like the footbridge case. That study, Greene says, marked the first time that anyone had correlated a pattern of neural activity with moral judgment behavior.

“I have the hypothesis that somehow the idea of pushing the guy off the footbridge in the second case is more emotionally salient,” Greene says, whereas throwing a switch is less so. In the switch case, he suggests, it may be that our utilitarian brain is doing the thinking. So, despite the fact that numerically the problem is the same, one life for five, we intuit different moral judgments because different brain functions are involved in thinking through each problem.

About eight years ago, he teamed up with Jonathan Cohen, director of the Center for the Study of Brain, Mind and Behavior at Princeton, and became the first neurophilosopher. And then, says Dartmouth’s Sinnott-Armstrong says that since Greene’s second publication,  programs in the empirical study of philosophy have emerged at universities across the country.

In 2002, Greene completed his doctorate in philosophy at Princeton and in 2006, he came to Harvard as an assistant professor. His corner office on the 14th floor of Harvard’s William James Hall overlooks much of the University, including the new site of the brain imaging center, which is still under construction. For now, he is conducting his experiments at the Massachusetts General Hospital imaging center, working with variations on the classic trolley problem.

Weighing a baby's life

"So it’s wartime,” Greene begins, presenting another problem, “and you’re hiding in the basement with some fellow villagers and your baby. The enemy soldiers are outside and they have orders to kill all remaining civilians, and if they find you they’re gonna kill you and your baby and everybody else.”

This is known as the “crying baby scenario,” and it is of particular interest to Greene because it offers a hard choice between emotional and reasoned responses. People disagree about which answer is correct, and many take a long time to reach any answer. Greene wants to show that in these cases, a part of the brain that is associated with competing impulses is more active.

“Your baby starts to cry,” he continues, “and if you don’t cover your baby’s mouth, the soldiers will find you and kill everybody. But if you do, then your baby will smother to death. Is it morally acceptable for you to do this?” he asks.  “It’s a horrible question,” he continues. “No one likes to think about it very much, and some people say, ‘I guess so’ and some people say, ‘No, no that would be wrong.’ In those difficult kinds of cases you see more activity in the part of the brain called the anterior cingulate cortex, which tends to become active when there are competing behavioral responses, and not just in a moral context.”

Greene’s hypothesis suggests that we have an emotional impulse to think it’s wrong to smother the baby, as well as a utilitarian impulse to weigh the number of deaths with each possible outcome. Moreover, different parts of the brain are at work in the emotional and utilitarian case.

A multi-faceted process

“So if you had to just sum it up,” says Greene, “I guess the overall lesson is that moral judgment is not a single kind of process … at least in my view not a single moral faculty or moral sense, rather it’s different systems in the brain in some cases competing with each other.”

Sinnott-Armstrong said Greene has “done more than anyone to show how neuroscience can illuminate traditional philosophical disputes.” His contributions have prompted “traditional moral philosophers [to] get much clearer about which questions they are asking, and that itself is a significant contribution to philosophy.” He has also “introduce[d] a new method for understanding moral judgments.” 

But how do cultural differences fit into Greene’s approach to moral intuition? “The trend,” says Greene, “is that at least with these trolley sorts of questions, people’s intuitions are surprisingly stable across cultures. But we know that there’s a lot of cultural variability in terms of people’s moral intuitions because people from different cultures have quite stark moral disagreements.”

Greene is careful to point out that his work does not suggest that we’re born with an innate set of moral judgments. “Just because something is in the brain doesn’t mean it’s hardwired,” he says. Just as we aren’t born speaking English or Chinese, we aren’t born with a fixed moral sense.

Morality and neurons

But in the end, all moral decision making is a neurological process of some kind, Greene says. “We’re used to thinking of ourselves as having brains, but not being brains. That is, we think of ourselves as having a soul … and it’s sort of separate from the physical brain stuff, and if there’s anything that your soul traditionally does, it makes moral judgments… . So, thinking of moral decision making as a physical brain process and as merely a physical brain process, that may be a shock for a lot of people. … But the more we understand,” Greene believes, “the less likely it seems that there’s something going on beyond the firing of neurons in certain complicated patterns.”

But what exactly do trolleys and neurons have in common? What might an out-of-control railcar reveal about our moral brains at large?

“I view these little trolley problems as like fruit flies,” says Greene, explaining that “they’re sort of simple little systems that are still surprisingly complicated. And they’re nice because you can bring them into the lab and poke at them and prod at them and they’re simple enough that you can study them and get meaningful results … but they’re complex enough that they capture something that’s interesting about the more complicated real-world moral dilemmas that we’re really interested in.” By understanding how our brain deals with lab-grown conundrums, Greene hopes we might learn how to better handle broader moral and social problems.