May 18, 2011 | 16
I witnessed a breakup yesterday in the middle of MIT’s vast Infinite Corridor—a hallway known for its heavy traffic and long stretch of straightness. Finals are upon the undergraduates, so perhaps tensions were a bit high for the young, failing couple. Something, however, had clearly pushed the girl overboard. Her boyfriend had fallen dramatically to his knees and as he wept heartfelt apologies for some crime or another, the girl stood with crossed arms, trying not to look at him. Then, as I passed, the angry young woman knelt and slapped him hard and loud across his face just before storming off down the Corridor.
I don’t know what happened between those two, but I felt bad for the guy. Seeing him cry so openly in public hurt me, in a small way. And when the slap connected with his wet cheek, the loud clap stung my face in a phantom sort of way. All that insult-to-injury—it hurt.
This, I believe, is empathy (though, as I’ve recently discovered, scientists can’t seem to agree on their own definition). We’ve all "felt" for someone else, whether that person is a stranger getting slapped in public or a close friend suffering through the loss of a pet. The empathy we experience can feel as real as if the pain were our own.
But empathy is failing on a pretty mass scale. It fails between Christians and Muslims. It fails between Israelis and Palestinians. Between Democrats and Republicans. Between Red Sox and Yankees fans. When it comes to conflict groups, empathy largely goes offline.
Because of this failure to empathize, MIT Saxelab neuroscientist Emile Bruneau has set his sights on not only locating empathy in the brain via controversial fMRI scans; he also hopes to find a way to quantify empathy.
The scanning room at MIT’s McGovern Institute for Brain Research is small, a bit like a cage. With only a row of computers and a few focused but tired graduate students toiling away on spreadsheets, it’s what I imagine a human terrarium might look like. And on the other side of the room’s single window is a large, round, glowing machine that looks like it could become self-aware at any minute and start sealing out oxygen and talking in a sinister, monotonous voice.
This machine is a functional magnetic resonance imaging (fMRI) scanner, and Emile Bruneau has signed up for several days’-worth of usage to run his current experiments. His first subject on the day of my visit was a young Israeli woman, dressed all in scrubs. As Bruneau led her into the scanning room to situate her and give instructions on what to do, the graduates back in the terrarium with me typed away on their laptops, presumably getting things ready.
All the subjects that day—all the subjects for this ongoing study—were Israeli and Palestinian volunteers, of random ages (over 18) and sex. One by one they marched into the scanner to get their brains’ pictures taken, each taking about 40 minutes (which, by the way, is a really long time to keep your head still).
The setup is pretty simple. Bruneau’s subjects lie in the fMRI scanner while short, 60-word stories appear before their eyes. The stories tell quick tales of another in emotional or physical pain—nothing too extreme. As a subject reads over the stories, a prompt appears after a few seconds asking her: How much compassion do you feel? She rates from 1 to 5, depending on the intensity of her admitted experience. Images of the subject’s brain (the fMRI measures blood flow, so the more active regions register on the images) are finally sent back to Emile, who watches from his computer.
This isn’t Emile’s first rodeo. He recently ran an almost identical experiment, only instead of limiting his subjects to members of well-known conflict groups, his subjects were random MIT students. The stories and prompts he showed them were identical to those used for the conflict groups, except without any details of the protagonist’s ethnicity. Whereas John, Jerry, or Diane might have been the main character of a prompt in the first, MIT-student run, characters like Ashraf, Fadwa, or Tal took center stage in this more focused round. Bruneau wanted to see whether or not a Palestinian subject would feel less empathy for an Israeli character, and vice versa.
Bruneau discovered with the first experiment what appeared to be concrete anatomical structures in the brain that responded to physical and emotional pain in others. As he monitored the young Israeli woman in the scanner, he pulled up some of the results from his previous experiments to show me what he meant.
"The first thing you can see is that there are distinct brain regions that are responding to these types of stimuli," he said, pointing to rather colorful brain scans (though, he qualified, the results don’t look so appealing straight out of the scanner). There were, in fact, distinct red and green areas on the scans. The regions in red, he said, represented how subjects responded to others in physical pain; the green areas represented their response to emotional suffering.
"The second thing you can see is that they’re completely separate from each other. So the brain regions that are responding to physical pain are very different than the brain regions that are responding to emotional suffering," he trailed off, leaning over a speaker connected to the scanner. The next round of stories are coming up when you’re ready, he told the Israeli woman.
These conflict-rooted empathy experiments haven’t been running too long yet, and he’s yet to dig into the imaging data, but Bruneau has a few expectations, a few questions. For instance, how do members of these conflict groups experience empathy?
"The prediction isn’t that you’ll get a completely different brain response in totally different brain regions," he explained. "The prediction is that you’ll get the same pattern of response, it’ll just be decreased so that there’s less of a response in those brain regions."
In other words, the empathy is there. It’s just on mute.
"One of the strengths of neuroimaging is that you can get a quantitative measure of the activity in different regions, and that’s kind of what we’re relying upon," Bruneau said.
If he succeeds in quantifying empathy, then perhaps, he says, there will finally be a way to measure whether or not conflict resolution programs are working. If a participant in such a program has "greater" empathic activity in his brain after a program, then, obviously, it’s working. If not, adjustments to the program would have to be made.
All in all, Bruneau really hopes to bring a little peace to a conflict-charged world.
But this is a lofty goal, and the methods are not without critics.
Glimpsing the Unglimpsable
Ten years ago, cognitive neuroscience was skeptical about trying to localize any kind of social process in the brain. The endeavor has been likened, by the harshest of critics, to a new wave of phrenology, the pseudoscientific idea that bumps on a skull indicate some kind of "brain map." The skeptics argue that such behaviors and processes, like empathy, can’t be pinpointed to one specific place—the processes emerge, rather, from complex network interactions in the brain.
Bruneau and his supervisor, Rebecca Saxe of Saxelab, both believe that this skepticism has all but died off, and that the whole localization versus network-distribution battle is moot.
"[Localization and distribution] aren’t really alternatives to one another. They might be alternatives to each other in the discourse of the literature, but they’re not alternatives in reality," Saxe responded to the issue. She silences the debate by drawing an analogy to the motor of a car. Sure, an engine works because many pieces work together in unison—in a network, if you will—but that’s not to say a carburetor doesn’t perform a specific job. Saxe emphasizes that this is an oversimplification, but it gets at her basic idea.
Even if localizing specific brain functions is at some point wholly agreed upon as a worthwhile endeavor, fMRI scanning has its critics, too. Perhaps the most colorfully demonstrated argument against the efficacy and reliability of these scanners came about in October 2010. Researchers stuck a dead salmon in an fMRI, asked him questions about human emotions, and measured his response. That is to say, the dead salmon had brain activity show up in his scans. The dead fish had thoughts on human emotion.
Of course, the fish wasn’t responding in any way, shape, or form to his prompts. He was dead. And a fish. This study only served to drive in the point that there are data corrections that must be taken into account for fMRI scans; the results can yield false positives. Regardless, neuroimaging, as some have astutely pointed out, is seductive to researchers, to the people who fund them, and, especially, to the rest of us. After all, it’s nice to see answers to difficult questions simply light up on a screen.
There’s clearly a debate, and probably no end to it in the near future. Matters of the brain are tricky business. I’m no neuroscientist, just a hobbyist, but both sides of the argument seem well-reasoned, well-researched, and well-intending. While it’s exciting to see those brain scans that Bruneau brought up on screen, there’s also something oversimplified about it all.
On the one hand, if he finds significant results, it could lead to groundbreaking methods for palliating some of the world’s problems. On the other hand, it can feel a bit alienating, a bit deflating, to think of one’s brain as simply a series of green and red regions. Then again, reductionism has a tendency for that sort of thing.
As I sat in my corner and watched Bruneau conduct his experiment, I saw one of the prompts flash on a screen near the neuroscientist. It read:
Farrah lives with her new husband, who is a bank manager in Baghdad. The couple often has friends over for dinner. One night, as Farrah was chopping vegetables, a friend asked her a question. When Farrah turned to respond and slipped with the knife. The knife cut a huge slice in her finger that went to the bone.
The writer in me should have been paying attention to the way the protagonist was being presented and the ethnically charged rhetoric—using the name Farrah repeatedly, the mention of Baghdad, for instance. Or the way the prompt was crafted to elicit several responses.
Instead, I clenched my teeth and felt a phantom knife slice down to my bone.
I wonder if the subject reacted the same way.
Images Credits: Stephen McCarthy
About the Author: Kristina Bjoran is a student in MIT’s Graduate Program in Science Writing, where she writes about technology, environmental studies, and nonhuman animal intelligence. During her down time, she volunteers with animal shelters, writes for nonprofits, and dabbles in photography and skydiving. Follow Kristina on Facebook and Twitter, and visit her page on MIT Scope.
The views expressed are those of the author and are not necessarily those of Scientific American.