The Fear Factor
Page 15
A fearful facial expression (here as posed by Paul Ekman) includes widened eyes, raised, oblique brows, and a grimace.
Abigail Marsh.
The rapid emergence and primitive control of human facial expressions is mirrored by the rapidity and primitiveness of how other brains respond to the sight of them. To appreciate this, let us now bestride our beam of light and tag along as it carries information about a fearful face deep into the brain of an observer.
In the nanoseconds after a person’s face registers fear, much of the light striking his or her face ricochets backward from it again, radiating outward toward every other creature in the vicinity and carrying information about that face with it. That is all it means for an object to be visible: that it reflects light back toward the viewer’s eyes. The regions of the face don’t release reflected light evenly, though—pupils swallow it hungrily, while bright white sclera fling it nearly all back. Variations in the density and direction of reflected light enable the light to carry with it a detailed recording of all the curves and colors of the face. Let’s choose one of the many available beams reflected from the sclera upon which to ride, and off we go, racing at 3 million meters per second toward the eye of a nearby human onlooker.
We reach it nearly instantaneously. After we cross the clear dome of the cornea, we pass through the pupil and into the prism of the lens, which twists us upside down and sideways as it focuses the incoming light into a sharp image. Then, after we sail through the clear jelly filling the eyeball, we make a soft, inverted landing on the fleshy retina at the back of the eye. Here we make an astonishing transformation: we are digitized. The information our light beam carries about the wide, white sclera where it originated is transformed into a digital blip of information by photoreceptor cells in the retina. Millions of these cells pulse when struck by bright white light, sending staccato messages down the optic nerve toward the brain.*
Our beam of light, now transformed into a nerve impulse, is racing away from the eye down the optic nerve on its way to the onlooker’s brain. Our speed has slowed but is still blisteringly fast by human standards, about 60 meters per second. As a result, mere fractions of a second after the onlooker has taken in the wide, white sclera—plus the brows and mouth—of a fearful expression, that information has set the onlooker’s brain aflame.
It’s hard to overstate the effect of seeing another person’s fear on a human brain. The sight changes patterns of activity in nearly every crevice of the brain, although not all at once. The first region to receive the message from the retina is a pair of evolutionarily ancient structures deep in the brain’s core called the superior colliculi. The colliculi are two backward-facing nubs of tissue perched like pert Barbie breasts atop the brain stem. Their role is to rev up a lightning-fast response to important visual information coming in, well before the person even has any conscious awareness of what was seen. Images processed in the colliculi aren’t sharp or detailed, but what they lack in precision they make up for in speed. Like a microscopic, warp-speed relay race, the colliculi pass the gist of the information carried by beams of light (“Lots of sclera! So much white!”) to new fibers that extend upward to an oblong mass of neurons perched in the center of the brain, the thalamus. We zip in milliseconds to this structure, which acts like the brain’s switchboard, taking in signals from dispersed areas of the brain and relaying them out again to other areas. When it receives the signal from the colliculi that the wide, white sclera of a fearful expression have been detected, the thalamus knows just where to send that information next—to the amygdala.
Findings presented in a 2016 article in Nature Neuroscience demonstrated for the first time that visual information about human fearful facial expressions is conveyed via this long-hypothesized ancient pathway. The researchers inserted electrodes directly into the amygdalas of eight adult humans to record activity there while the researchers showed them pictures.* They found that a mere seventy-four milliseconds after a fearful face flashed across a computer screen, the electrodes began buzzing with activity, signifying that the amygdala had already received information about the rough contours of the face and begun to generate a response. This is far too fast for the information to have arrived in the amygdala via any pathway other than the rapid, ancient route we have just traveled through the colliculus and thalamus. And here’s the wild part: no other facial expression that we know about gets passed along this same privileged, speedy route to the amygdala. Not resting faces, not happy faces, not angry faces. Just fear. The mystery is: why?
Let’s follow our transduced light beam for a moment more before digging into this mystery, which is itself deeply intertwined with the mystery of human altruism.
Upon exiting the thalamus, we arrive first in the amygdala’s lateral nucleus, one of several semi-separate clusters of neurons within the amygdala, each of which serves a distinct role. The lateral nucleus is a sort of foyer to the amygdala where most incoming information arrives. Here we may be forced to watch helplessly as the message carried by our light breaks up, caroming in dozens of different directions simultaneously through the rest of the amygdala, then outward through the rest of the brain, as multitudes of neural cavalry are rallied to respond to what has been seen. The hubbub of activity in the amygdala following the perception of a fearful face is much greater than what follows the perception of any other expression. This is true even when the fearful expression is mostly obscured, leaving only the sclera visible. It’s true even if those sclera are presented so quickly that the viewer has no conscious awareness of having seen anything at all. Dartmouth College professor Paul Whalen and his colleagues once demonstrated this by flashing just the wide, white sclera of fearful facial expressions on a plain black background to brain imaging study participants for a mere seventeen milliseconds—far too quickly to be consciously detected. They found that the amygdala still burst into a furious volley of activity—much more than when only the sclera of neutral expressions were presented. This remarkable degree of sensitivity shows that others’ fear is unusually important information to the amygdala. But why?
For quite some time the thinking was that fearful expressions are important because they tell viewers that they should be fearful too. A person expressing fear is clearly afraid of something—a snake, a gun, the edge of a cliff. The resulting facial expression, according to this story, serves as an alarm signal telling anyone else in visual range that they may need to flee or brace for danger.
It’s not an implausible explanation. Most social species use alarm signals like special calls to warn others around them of danger. Sending such calls is actually considered a form of altruism, as callers risk drawing predators’ attention to themselves to warn others of danger. And just as theories of kin selection and reciprocity would predict, risky alarm calls are most likely to be used to alert family or other social group members. The auxiliary benefits of these calls extend far beyond the caller’s family or even the caller’s own species, though. Many species benefit from the alarm calls of even distantly related species. Birds can recognize the alarm calls of other local species of birds and even squirrels. Tropical toucanlike birds called hornbills even distinguish between and respond appropriately to the two distinct alarm calls that neighboring Diana monkeys use to warn against different types of danger (leopards versus eagles), as though they have learned the monkeys’ language of fear.
Do fearful expressions in humans serve a similar purpose as these calls? Many have argued or assumed that the amygdala’s robust response to fearful expressions is proof that they do. As a rule, the amygdala does respond rapidly to sensory events in the world that portend danger—the rippling eddy of a snake, the click of a gun being cocked, the feel of the wind along a cliff. The amygdala can learn very quickly, sometimes after a single trial, to link cues like these to incipient harm. Thereafter, when these cues are detected, cells within the amygdala fire furiously, sending urgent messages out to the rest of the brain that danger is near. My mother can thank her amygdala f
or the fact that she once found herself leaping into a frantic and slightly embarrassing stationary panic in front of the neighbors after a harmless garter snake slithered across our driveway. She had just recently returned home from a trip to the Amazon rainforest, where her tour group had nearly stepped on a deadly fer-de-lance lying across their path. My mother’s amygdala had not forgotten how close she had come to danger. The coordinated volley of firing in the amygdala in response to danger is central to the felt experience of fear, as we know from studying patients like S.M. who lack both an amygdala and the ability to experience fear and from studying psychopaths in whom both the amygdala and the experience of fear are stunted.
So yes, it’s certainly possible that amygdala responses to fearful expressions represent a learned response that these expressions signal the presence of danger. But there are also problems with this explanation. First, it’s unlikely that this is the primary function of fearful expressions, given the impracticality of a visual alarm signal. Eyes, it seems almost too obvious to say, see only what they’re looking at. What if you’re looking in the wrong direction, or blinking, or sleeping, when the alarm goes off? There is a reason why fire alarms don’t take the form of a little flame symbol silently lighting up in the ceiling. Ears and noses, by contrast, are always open and picking up information coming in from any direction. As a result, in most species, alarm signals take the form of barks and squeals or bursts of pheromones, not visual cues. For the same reason, the fearlike facial expressions of other primates don’t really function as alarm signals. Instead, they are used to signal submission and appeasement—to inhibit others’ aggression.
Amygdala responses to fearful expressions are also quite different from responses to other expressions that clearly signal threat. Angry facial expressions provide an interesting contrast. When someone is staring at you with their eyes narrowed, their brows lowered, and their teeth bared, this is clearly a threat. Anytime you see a face like this, an aggressive attack may be imminent. The face of the man who broke my nose in Las Vegas contorted exactly this way right before he hit me. But the amygdala normally doesn’t respond to angry facial expressions at all. Angry faces actually generate even less of an amygdala response than a neutral resting face. And the amygdala’s response to threatening scenes, like images of mutilated bodies, also looks different than its response to fearful faces. When the researchers who measured activity in the implanted amygdala electrodes showed their subjects images like these, they found no comparable rapid response. This almost certainly means that information about these scenes had arrived in the amygdala via a different path.
Yet another problem with the “threat response” theory is that it has difficulty explaining why damage to the amygdala impairs not just people’s ability to respond appropriately to fearful expressions but their ability to even identify them—to come up with a name for what the expresser is feeling. When S.M. sees a fearful face, it isn’t as though she knows what to call it but fails to show appropriate signs of fearful avoidance or vigilance in response. It’s that she sees it and is mystified by its very meaning, like a color-blind person searching for a number in a featureless array of brown dots.
Psychopaths’ seeming blindness to others’ fear can also be striking. I am still haunted by a story once related at a conference by my friend and colleague, the psychopathy researcher Essi Viding, at University College London. She was testing a psychopathic inmate in an English prison and had shown him a long series of emotional faces. He was among the subset of psychopaths who are completely blind to others’ fear: he got every single fearful expression wrong. Not once did he recognize the wide eyes, oblique brows, and grimace of a fearful face as signifying fear. He knew he was performing badly too. When he got to the final fearful expression in the set and yet again failed to identify it, he mused aloud, “I don’t know what that expression is called. But I know that’s what people look like right before I stab them.”
Remarkably, this psychopath was able to recall having seen similar expressions before—and even to pinpoint the circumstances in which he’d seen them. But he was unable to discern that this particular and familiar combination of features, even in an obviously frightening situation, signified fear. How can this be explained? Not by the “threat response” theory.
There is another quite distinct (although not mutually incompatible) explanation for all of these findings, which is that amygdala responses to fearful expressions represent not a response to “threat” but rather a deep, atavistic form of empathy.
When a light-borne message arrives in the amygdala that the wide, distressed eyes and grimace of a fearful face have been detected, the cascade of neural firing that ensues in this structure may actually reflect a simulation of the interior state of the expresser—almost like an internal translation of the other person’s fearful state. It’s this simulation that allows the perceiver to understand and put a name to the expresser’s state, but leaves those without functioning amygdalas drawing a blank. It’s this simulation that causes faint whispers of fear to cascade down from the amygdala to a nub of brain tissue called the hypothalamus and from there outward through the rest of the body, causing most people’s hearts to beat a little faster and their palms to sweat a little more in response to seeing another’s fear—yet another response, not incidentally, that S.M. and psychopaths fail to show.
If all this is true—if a tiny blip of digital information carried on a beam of light can create an echo of a fear response in another person—it means that amygdala responses to fearful facial expressions represent a true conjoining of the interiors of two human brains. This would be a monumental thing. The ability to internally re-create another person’s emotion, and thereby understand it, is a basic but essential form of empathy. This form of empathy is critical to the capacity to generate still more profound social responses, like caring that another person is frightened or distressed, and wanting to make that person feel better.
This isn’t such a far-fetched possibility. A similar sort of empathic response has already been identified in various parts of the brain in response to pain. Dozens of brain imaging studies have now shown that the sight of another person in pain results in increased activity in a constellation of brain regions called the pain matrix. These regions include cortical regions like the mid-cingulate gyrus and anterior insula as well as deeper, subcortical regions that are also active during the personal experience of pain. The uncanny overlap in the regions that become active both when experiencing or witnessing—or even imagining—another person’s pain strongly suggests an empathic response.
Even stronger support for this possibility comes from a clever brain imaging study reported in 2010 by Tania Singer, Grit Hein, Daniel Batson, and their colleagues, who examined empathic pain responses in sixteen Swiss soccer fans. All the fans were selected for being impassioned supporters of their local team. The researchers wanted to know how they would respond to the sight of pain being inflicted both on fellow fans of the local team and on fans of a rival team, using—you guessed it—electric shocks.
After each soccer fan arrived for the study, he was positioned in the MRI scanner by the researchers, who then taped customized electrodes to the back of his hand. Once the scan began, so did the shocks. The researchers measured the subjects’ brain activity as electricity coursed through the electrodes and across the skin of their hands. The shocks varied in intensity, with some being very mild, and others being more painful. When the researchers analyzed the subjects’ brain data afterward, they found, as expected, that activity in the anterior insula, a key component of the pain matrix, ratcheted steadily upwards as the intensity of the pain increased. The insula lies deep beneath the temples on either side of the head and is thought to encode the emotional significance of unpleasant body sensations. When those parts of the insula are active, in other words, it signals that what’s happening feels bad. What the researchers wanted to know was how this same area would respond when the subject watched allies
and rivals experiencing pain. Would it signal that what was happening to other people also felt bad?
During the study, each subject was flanked by two strangers, and all three of them were wired up to electrodes. It must have been quite a squeeze, with three grown men lined up side by side in the cramped confines of a scanner room, all of their hands positioned to be visible to the subject lying flat on his back in the scanner, peering out of it through an angled mirror. On one side of the subject sat the ally, who the subject knew was a fan of his own team. On the subject’s other side sat a fan of a rival team. As the subject watched from inside the scanner, both the ally’s hand and the rival’s hand were also subjected to electrical shocks of varying intensities. Imagine it: You’ve just met a stranger, spoken to him for a few minutes, and know that he shares your love and loyalty for your favorite team. Now imagine watching his hand twitch and jerk as electric shocks jolt through it. Would you cringe with discomfort? Twitch slightly yourself? Batson and Singer’s findings suggest that you might. As subjects watched the ally being shocked, activity increased in the same region of the anterior insula that was active when they experienced pain themselves, just as you would expect if the subject were simulating the ally’s pain. Remarkably, though, subjects’ response when they saw their rival shocked was quite different. As this stranger’s hand twitched and jerked in response to the shocks, the subjects’ insulas were nearly silent.
From Batson’s prior research, we know that many participants in a study like this will not only experience concern for a stranger being shocked but be willing to actively help the stranger by taking on extra shocks themselves if need be. Singer and Batson again found this to be true. When given the opportunity to take on half of a stranger’s remaining shocks, many participants volunteered to do so. But again, this was largely only true for their allies. When the rival received the shocks, participants were much less likely to offer to help. More, the participants’ willingness to help fellow fans rose and fell in tandem with activity in their anterior insula. The more the insula responded empathically to a fellow fan’s pain, the more likely it was that help would be extended.