Perhaps even more impressive is what happens even earlier, at three months. At that tender age, if infants have performed similar actions, they are more likely to understand the goals of others’ actions. At three months, infants don’t have good motor control, they cannot yet reach and grasp reliably, and their hands flail about. The clever experimenters put mittens with Velcro on the baby’s hands and a toy in front of the infant. Eventually, with enough flailing, the Velcroed hand would catch the toy. The infants who had had practice “grasping” objects in this way anticipated the viewed reaching and grasping actions of others more reliably than infants without practice grasping.
This is remarkable evidence that infants can understand the intentions behind the actions of others. Not all intentions and actions, of course, but reaching for an object is an important and common one, and there are undoubtedly others. Understanding others’ intentions comes about in part because of experience enacting similar actions with similar intentions. Moreover, as we shall see next, it has become clear that the very structure of the brain is primed for understanding observed action, through the mirror neuron system.
MIRROR NEURONS
In the late 1980s, a group of neuroscientists in Parma, Italy, led by Giacomo Rizzolatti made a surprising discovery. They implanted tiny electrodes in individual neurons in premotor cortex (inferior frontal gyrus and inferior parietal lobe) of macaque monkeys that allowed them to record activity in single neurons in animals who were moving about as they normally do. They found single neurons that fired when the monkey performed a specific action, like grasping or throwing. What was remarkable was that the exact same neuron fired when the animal saw someone else, in this case, a human, perform the same action. They called these remarkable neurons mirror neurons. Mirror neurons unite doing and seeing for specific actions. This extraordinary discovery means that action and perception are joined automatically by specific individual neurons without any mediation whatsoever. Different actions are encoded by different neurons: for monkeys, grasping, throwing, tearing. You can watch the action and listen to the simultaneous firing of these neurons online. More generally, the finding suggests that action mirroring, sometimes called motor resonance, underlies action understanding. Seeing is mapped to doing and doing is mapped to seeing. I understand what I see you doing because an echo of the doing resonates in my own action system. Of course, it’s just an echo; I’m not actually doing what I’m seeing, which is a good thing. Otherwise, we’d be caught in an unending cycle of imitation. Mirror neurons underlie the understanding part of imitation, but not the doing part.
Naturally, these findings, now replicated many times, have generated enormous excitement. Overinterpretation of tantalizing findings like these is inevitable. Might mirror neurons underlie imitation, learning, and memory? The research group in Parma has gone to great effort to explain that seeing is not imitating and that understanding is not doing. If it were that simple, we’d all be expert pianists or basketball players or acrobats. Motor resonance, however, is real: that is, seeing action causes associated motor regions of the brain and even associated muscles to activate.
It’s problematic to perform these experiments in humans. We don’t simply implant electrodes in individual neurons in other humans. There are cases, however, when recording from single cells in the alert human brain is crucial for people’s health and well-being, for example, for people with intractable epilepsy. Epilepsy can often be controlled by destroying the brain tissue that seems to initiate seizures, but neurosurgeons first make sure the brain locations are not involved with core functions like speech. The way to find out is by implanting electrodes in the suspect areas, and sometimes in other areas, where the electrodes would cause no damage. Studies recording from single cells in those patients have found evidence for mirror neurons in multiple parts of the human brain, for example, individual neurons that responded when people observed or performed actions and also when people viewed or expressed facial emotions.
The actions of bodies are qualitatively different from the actions of other objects. A crucial difference: bodies are self-propelled, which means bodies can perform gravity-defying actions, like jumping up in the air all by themselves, feats that baseballs and leaves cannot perform. Even small children do a decent job telling whether a path of motion is from an animate being or an inanimate object. Of course, animacy is far more than paths of motion, and even small children understand that. Yet, it is significant we and small children can make a good guess as to whether something moving is animate or not simply from an easily perceivable path of motion. It is sometimes surprising how much deep essential qualities like animacy are apparent from superficial perceptual features. Earlier we saw that objects can be recognized by their contours. Other examples to come.
MOTOR RESONANCE
Just as infants understand viewed actions better if they have performed the same actions, not surprisingly, so for adults. Our experience performing specific actions modulates our perception of the same actions performed by others. In an experiment that has generated smiles, excitement, and even some controversy, experts in capoeira, experts in ballet, and nonexperts viewed videos of standard movements in capoeira and ballet while in a scanner that measured their brain activity. Brain activity in a network involved in the mirror system (premotor cortex, intraparietal sulcus, right superior parietal lobe, and left posterior superior temporal sulcus) showed more activation when observers watched the movements for which they were experts.
The broader implication is by now a familiar one: we understand actions that we view by simulating the actions in our bodies, by embodying the perception. Many names for more or less the same phenomenon: motor simulation, motor resonance, embodiment. There are even more.
Motor simulation has implications beyond understanding action. It affects our predictions and expectations about action, for example, whether a basketball will make the basket. In one study, professional basketball players, sports reporters, and professional basketball coaches were asked to predict whether free throws would make the basket. Coaches and sports reporters have extensive outsider, that is, visual, experience watching basketball, and from many different points of view. Basketball players have that visual experience, but they also have the insider view. They know what it feels like to shoot a basket, and most likely they have developed good intuitions about which shots they take themselves will make the basket. Professional basketball players are so practiced at shooting baskets that they have often been called shooting machines.
You’ve guessed the results. All three groups, coaches, reporters, and players, were impressively good at predicting which balls would land in the basket. That said, the players were far better. Their extensive insider knowledge enabled professional basketball players to predict shots better than coaches or sports reporters. The experimenters stopped the videos at varying intervals from the beginning of the shot until the ball was very close to the basket. What was especially impressive is that the players’ predictions were superior even before the ball left the player’s hands! This suggests that the players had insider motor understanding of the body kinematics underlying basketball shots, and that this understanding allowed them to better predict the outcomes of the actions. Players have more motor experience than coaches and sports reporters and that motor experience enables better predictions. Along with other evidence and more to come, it suggests that we map body action that we see onto our own body’s action system. Perception of action acquires meaning through motor understanding. Experts with more articulated motion systems perceive more meaning in what they see.
Back in the 1970s, a Swedish psychologist, Gunnar Johansson, dressed people in black and attached small lights to their major parts and joints, head, shoulder, elbow, wrist, hips, and so forth. He then filmed them as they performed a common set of human actions, walking, running, and dancing. This paradigm, using point-light videos, has been adopted, adapted, and simulated many times since. You can find many captivating exa
mples online. A static image of any of the point-light bodies is unidentifiable; it looks like a random collection of dots. But once the set of dots is set in motion, you can immediately see that it’s a human body, you know if it’s walking or running or dancing, you know if it’s a man or a woman (by the ratio of the shoulders to the hips), you can tell if it’s happy or sad, energetic or tired, heavy or light.
More recently, a group of researchers used that paradigm to ask how well we can recognize individuals from point-light videos. They brought pairs of friends into their laboratory, dressed them in black, attached lights to their heads and joints, and filmed them dancing, running, boxing, walking, playing Ping-Pong, and more, altogether ten different activities. Some months later, all participants were invited back to the lab to view all the videos. For each video, they were asked to identify the specific person moving. Perhaps not surprisingly, people were pretty good at recognizing their friends from their friends’ movements but poor at recognizing strangers. It was far easier to recognize individuals from the videos depicting more vigorous activities like dancing, jumping, playing Ping-Pong, and jumping than from the videos showing walking and running. The next finding is truly surprising. Participants were best at recognizing themselves! Most of us, and certainly the participants in this experiment, don’t spend a lot of time looking at our movements in the mirror unless we are dancers or models or yoga practitioners. How is it that we are best at recognizing ourselves, people we never or rarely see playing Ping-Pong or jumping, better than we are at recognizing our friends, whom we’ve presumably had extensive experience watching in motion? As before, the mirror system, motor resonance, seems to underlie that impressive ability. The theory goes like this. As participants watched the videos of people in action, their mirror systems resonated to the actions they were seeing as if they were trying the movements on for size. When they watched videos of themselves in action, the movements fit perfectly, they felt right, felt natural, felt like themselves.
COORDINATING BODIES
Birds flock, fish school, troops march. Bees gather nectar, ants build nests; basketball players coordinate on the court; boxers, on the mat; improvisors, on the stage. Commuters race every which way through Grand Central. For the most part, there are no collisions and there’s no one directing the traffic. There are so many ways that organisms rapidly coordinate their behavior with each other and so many reasons for the coordination. The mere presence of others affects our behavior, without any need to coordinate. You’re sitting alone in a waiting room or on a seat in a train or in line to buy a ticket. A total stranger arrives and does the same, sits down across the aisle or stands behind you in line. Unless you are completely distracted, say, by the smartphone at your fingertips or near your ear, you can’t help but be aware of the presence of the other, and that presence affects your behavior.
In situations like those, standing in line, sitting in a waiting room or on a train, you and the stranger are performing the same action at the same time in the same space. Providing that there is plenty of room for each of you, your actions do not have to be coordinated. If the train or waiting room is crowded, you might have to coordinate, making room for each other and each other’s belongings. Walking down nearly empty streets doesn’t require much coordination with others; nor does clapping at the end of a performance. Yet, remarkably, pedestrians tend to synchronize their walking and audiences their clapping.
Presumably, like birds flying in flocks, the synchronization of the group organizes and eases the actions of the individuals. Since my walking or clapping is in sync with others’, I can attend less to mine. In the human case, perfect strangers fall into rhythm. Rhythm is deeply embedded in our bodies, in our hearts, our breathing, our brains, our actions—walking, talking, thinking, dancing, sleeping, waking—our days and nights. Our rhythms organize and synchronize our bodies and come to organize and synchronize our bodies with the bodies of others.
The games we play with babies practice those skills, although that’s probably not why we play them. The baby says “ahh.” We say “ahh.” The baby says “ahh.” We take turns doing the same thing. Later we change our responses slightly, they say “ahh,” we say, “ahh ahh.” We roll a ball to baby, baby rolls it back to us. We clap together or in alternation. Our play is unintentionally training the elements of joint action: synchronization, turn-taking, imitation, entrainment, joint attention, joint understanding. And we—and they—just thought we were having fun. We were. There is something so satisfying about doing something together, in sync.
In humans, coordination quickly turns into cooperation. As early as fourteen months, when a child sees an adult trying to get an object out of reach but close to the child, the child will hand the object to the adult. Both the social understanding and the social behavior are remarkable, all the more so because they don’t require language or any explicit coordination. Other primates, monkeys and great apes, can be induced to work together to achieve a joint goal. The standard laboratory task is jointly pulling on separate ropes to bring some nuts or bananas in reach. Elephants cooperate, as do dolphins, both often with humans, just like dogs. Indeed, research in Tomasello’s lab has shown that cooperation is the origin of moral behavior; we need to work together, but then we must, or rather should, split the rewards. When small children are given more than their share, they share their share with the others.
At the other end of the continuum of joint action are tasks that require continuous and continuously changing coordination. To study this kind of coordination, we brought pairs of students who had never met each other into a lab room. On a table was a stack of parts of a TV cart and a photo of the completed cart. The pairs were asked to assemble the TV cart using the photo as a guide. We’d done many experiments on TV cart assembly by then, so we knew that students could do this individually, even without instructions. We even began to think this simple experiment was an important part of their undergraduate education.
Sure enough, the pairs were successful at assembling the TV cart. They did it correctly and efficiently, if each pair differently. Unlike walking or clapping, assembling the TV cart required partners to work together while performing different actions that were the components of each assembly step. The pairs spontaneously assumed different roles, usually implicitly, without even speaking. One would take the role of the heavy lifter and the other, the role of the attacher. The heavy lifter would hold a large part steady so that the attacher could connect another part to it. Each step in the assembly was more efficient with both partners, and to accomplish each step, the partners had to perform complementary actions. What was fascinating is that so much of the coordination happened without explicit negotiation, without words, even though the assembly actions were asymmetric and had to be done together. What’s more, each partner knew what the other needed to do, often anticipating the partner’s next action. The heavy lifter might see that the attacher would soon need a specific part and hand it to the attacher. When the attacher had the next part in hand, the heavy lifter would position the base so the attacher could easily attach the part. And so on. A kind of dance.
It seems amazing that this intricate set of interactive actions can happen almost wordlessly, without explicit organization. But on deeper reflection, perhaps it should not be so surprising. An orchestra needs a conductor, but a string quartet does not. Jazz improv can be beautifully coordinated as can comedy improv, without scores or scripts or a leader. At the core of joint action is joint understanding, shared knowledge of the goals and subgoals of the task and the procedures needed to accomplish it. For the TV cart, the procedures are a sequence of actions on parts, placing each part in turn in the correct configuration and attaching it with the appropriate means of attachment. The joint understanding of the goals of the task resides in each partner’s mind. In fact, people have numerous event schemas in their minds, representations of the sequences of actions on objects needed to accomplish a range of ordinary tasks, like making a bed or doing the
dishes or assembling a piece of furniture. These representations allow people to interpret ongoing action, to predict what happens next in ongoing action, and to generate step-by-step instructions to accomplish the tasks.
Peering inside the brain can reveal some of the processes that track joint action. Both electroencephalogram (EEG) and functional MRI (fMRI) research shows that participants keep their joint task active in the brain as well as in the mind. Surprisingly, partners keep each other’s task in mind even when doing so interferes with their own performance, slows them down, and makes them more prone to error.
Although the representation of the task resides in each partner’s mind, the set of procedures to carry out the task relies on objects and partners in the world. Participants have to keep in mind the overall goal and procedures and use that to guide their own actions. To prepare their own actions, people need to monitor each other’s actions as they proceed step-by-step. This means that to collaborate, partners must share and maintain joint attention in real time. For this kind of ongoing collaborative task, joint attention doesn’t necessarily mean joint gaze. Think of a duet between a pianist and a violinist; their eyes are looking at different scores and their hands are playing different instruments. Their joint attention is on the music they are creating together. Rhythm, the most fundamental requirement for minimal joint action, is also the foundation for maximal joint action.
Conversation, too, requires complementary coordination and on many levels. Importantly, conversation partners collaborate on creating meaning, and much of that collaboration is direct, deliberate, and even deliberated. In actuality, conversation partners do not just coordinate the content and the timing of the conversation, they also coordinate aspects of their behavior that do not at first seem relevant to the conversation. They coordinate their actions, leaning forward to take the stage or backward to yield the stage, crossing or uncrossing their legs. They mimic each other, a phenomenon known as entrainment. They adopt each other’s words and phrases, even accents. They copy each other’s facial expressions, eye gaze, and body movements. These seemingly irrelevant behaviors turn out not to be irrelevant; they serve as “social glue,” showing and promoting mutual understanding, thereby enhancing communication and cooperation. If we use the same words, if we make complementary actions, we are understanding each other. And we like each other more.
Mind in Motion Page 3