by Neil Clarke
Remember that arm whose potential possible number of movements exceeds the number of atoms in the universe?
In 2002, the mathematician Emanuel Todorov came up with a theory in an attempt to answer this question. Basically, he argued that motor control is really an optimization problem for the brain. Optimality is defined by high-level performance criteria such as maximizing precision, minimizing energy consumption, minimizing control effort, and so on.
In the optimization process, the brain relies on the processing powers of the cerebellum. Before the commands for movement have reached muscles, the cerebellum predicts the results of anticipated motion, and then, combined with real-time sensory feedback, helps the brain evaluate and coordinate the motor commands.
A simple example: when ascending or descending a set of stairs, we will often stumble due to miscounting the number of steps. If feedback-based adjustments are made in time, we can recover and not fall. Feedback, of course, is often noisy and involves a delay.
Todorov’s mathematical model is compatible with all known evidence concerning the neural mechanisms of motor control and can be used to explain all kinds of behavior phenomena. Given some physical parameters, it’s even possible to predict the resulting motion using his model: for instance, how an eight-legged creature would jump in Pluto’s gravity.
Physics engines based on his models are used by Hollywood to produce naturalistic movements for avatars in virtual environments.
By the time I was in college, the Todorov model was already treated as textbook authority. Experiment after experiment provided more evidence that it was correct.
And then one day, Mr. Lu and I discussed Balin.
After I left home for college, he and I had kept in touch via email. He was like an oracular AI from whom I could get answers for everything: academics, awkward social situations, even relationship advice. We wrote long emails back and forth discussing questions that must have seemed ridiculous to anyone else, such as “would an out-of-body experience engineered by technology violate religion’s claim on spirituality?”
By an unspoken agreement, both of us avoided talking about my father.
Mr. Lu told me that Balin had been sold to another family in town, a nouveau riche household which was often mocked for conspicuous acts of consumption that appeared ridiculous in the townspeople’s eyes.
I had known that Father’s business had run into a rough patch, but I hadn’t imagined that he would be so short on cash as to consider selling Balin.
I shifted the topic to the Todorov mathematical model, and a new thought struck me. Balin was capable of imitating movements with perfect precision. Suppose we had him perform two sets of identical movements: one through subconscious imitation and the other by his own will; do these two sets of movements go through the same process of motor control?
Mathematically, there was only a single optimal solution, but was there a difference in the way the optimal solution was arrived at?
It took Mr. Lu three days to get back to me. Unlike his usual free-flowing, loquacious style, this time he wrote only a few lines:
I think you’re asking a very important question, one whose importance perhaps you don’t even realize. If we can’t distinguish between mechanical imitation and conscious, willed movement at the level of neural activity, then the question is: Does free will truly exist?
I couldn’t sleep that night. I spent two weeks designing the prototype experiment, and spent even more time studying the feasibility of my proposed study, soliciting feedback from my mentors and other professors. Then I submitted my project for approval.
It wasn’t until everything was ready that I realized that this experiment, one seeking to address a “fundamental question,” lacked a fundamental, required component.
I had no choice but to break my promise to myself and go home.
I’m going just to get Balin, I reminded myself again and again. Just for Balin.
Just like how A leads to B. Simple, right?
I once read a science fiction novel called The Orphans, which was about aliens who had come to Earth. They could imitate the appearance of specific humans and pass as human in society, but they couldn’t perfectly capture the characteristic ways their targets moved or the subtleties of their facial expressions. Many aliens, exposed as frauds, were hunted down by humans.
In order to survive, the aliens had to study how humans communicated via body language. They pretended to be abandoned orphans, and, once taken in by kind-hearted families, proceeded to use the opportunity of living together to imitate the mannerisms and expressions of their adoptive parents.
To the parents’ surprise, their children became more and more like them. And once the alien orphans decided that they had learned enough, they killed their father or mother, took on their appearance, and took their place. The scene where an alien killed his father and took his mother as his wife was unforgettable.
Though it became harder to tell aliens apart from humans, people finally discovered the fundamental difference between the extraterrestrials and humans.
Although the aliens were able to imitate human movements with perfect fidelity, they lacked the mirror neurons unique to human brains, and thus were unable to intuit the emotional shifts occurring behind human faces or to experience similar neural activation patterns in their own minds. In other words, they lacked the capacity for empathy.
And so humans devised an effective means to detect aliens: bring harm to those closest to the disguised aliens and observe the aliens for signs of pain, fear, or rage. The test was called “the stabbing needle test.”
The story’s lesson seems to be: humanity isn’t the only species in the universe that has difficulty relating to their parents.
Mr. Lu knew everything about Balin. He thought of the paoxiao as an example of overdevelopment in the mirror neuron system. He was fascinated by Balin, but he disapproved of the way we treated him.
“But he’s never resisted or even wanted to run away,” I used to counter.
“Overactive mirror neurons lead to a pathological excess of empathy,” he said. “Maybe he just couldn’t tolerate the look of abandonment in his tormentors’—your—eyes.”
“I guess that could be true,” I said. “I must be an example of underdevelopment of mirror neurons.”
“Cold-blooded, one might say.”
But when Mr. Lu took me to find Balin on my return, I realized that I wasn’t the most cold-blooded, not by far.
Balin was naked, his body full of bruises and lacerations. Thick, rusty chains were locked about his neck and shackled his arms and legs. He was shut inside a tiny brick-and-mud enclosure, about five feet on each side. The interior was dim, and the stale air saturated with the gag-inducing stench of excrement and rotting food. He was thinner than I remembered. Flies buzzed about his wounds, and the outlines of his skeleton poked from under his skin. He was like an animal about to be sent to the butcher’s.
He looked at me, and there was no reaction in his eyes at all. It was just like the first time we had met, that summer night when I was thirteen.
“They had him mirror the movements of animals . . . mating—” Mr. Lu was unable to continue.
Memories of the past flood into my mind in a flash.
I have no recollection of what happened next. It was as though I had been possessed by some spirit, and I moved without remembering wanting to move.
According to Mr. Lu, I rushed into the house of Balin’s new owners, and grabbed the Pomeranian beloved by the family patriarch’s daughter-in-law. I opened my jaw and held the neck of the whimpering creature between my teeth.
“You said, ‘If you don’t let Balin go, I’m going to bite all the way through,’” Mr. Lu said.
I spat on the ground. Though I didn’t remember any of it, this did sound like something I would do.
Mr. Lu and I rushed Balin to the hospital. As we were preparing to leave, Mr. Lu stopped me. “Do you want to see your father?”
&
nbsp; That was how I found out that my father had been hospitalized. Once in college, I had had almost no contact with him, and gradually, I had stopped even thinking about him.
He looked about ten years older. Tubing was stuck into his nostrils and arms. His hair was sparse, and his gaze unsteady. A few years ago, when Pu’er tea was all the rage, he had gambled with all his chips and ended up as the last fool to buy in at the height of the mania. He was stuck with warehouses full of tea leaves as the price collapsed, and ended up losing just about everything.
As he looked at me, I noted that his expression reminded me of Balin’s, as though he was saying, I knew you’d be back.
“I . . . I’m here for Balin,” I said.
My father saw through my facade and cracked a smile, revealing a mouthful of teeth stained yellow by years of smoking.
“That little gremlin? He’s much smarter than you think. We all thought we were controlling him, but sometimes I wonder if he was controlling all of us.”
I didn’t know what to say.
“It’s the same way with you. I always thought I was the one in charge. But after you left, I realized that you had always held a thread whose other end is looped around my heart. No matter how far away you are, as soon as you twitch your fingers, I suffer pangs of heartache.” My father closed his eyes and put his hand over his chest.
Something was stuck in my throat.
I walked up to his bed and wanted to lean down to embrace him. But halfway through the motion, my body refused to obey. Awkwardly, I clapped him on the shoulder, straightened up, and walked away.
“I’m glad you’re back,” my father said from behind me, his voice hoarse. I didn’t turn around.
Mr. Lu was waiting for me at the door. I pretended to scratch my eyes to disguise the emotional turmoil.
“Do you think fate likes to play jokes on us?” he asked.
“What do you mean?”
“You wanted to escape the route your father had paved for you, but in the end, you ended up in the same place as me.”
“I think I’m coming around to your way of thinking.”
“What’s that?” he asked.
“No one knows what will happen in the future.”
We’ve failed again.
The original premise is very simple: Balin’s hypertrophic mirror neuron system makes him the ideal experimental subject because his imitation is a kind of instinct. Thus, his movements during imitation ought to be free from much of the noise and interference found in human motor control due to conscious cognition.
We use non-invasive electrodes to capture the neural activity in Balin’s motor cortex as he’s imitating a sequence of movements. Then we have him repeat the sequence under his own will and use motion capture to ensure that we get a perfect match between the two sets of movements. Mathematically, that means that the two sequences are indistinguishable; they are the same motion.
Then, by comparing the two sets of neural patterns captured during the process, we can find out if the same neural signals were activating the same regions of the motor cortex in the same sequence and with the same strength.
If there are differences, then the Todorov model, accepted as gospel, will have been revealed to be seriously lacking.
But if there are no differences, the consequences will be even more severe. Maybe human beings are doing nothing but imitating the behaviors of other individuals, and are only operating under the illusion of free will.
No matter what, the result of the experiment ought to be earth-shattering.
Yet, the experiment was a failure from the very beginning. Balin has refused to look into anyone’s eyes, and has refused to imitate anyone’s movements, including me.
I can guess the reason, but I have no solution. My team has vowed to solve the secret of human cognition, yet we can’t even heal the psychic trauma inflicted on a mere “primitive.”
I thought of the idea of using virtual reality. Situating Balin in an environment completely disconnected from the reality around him may help him recover his normal habits.
And so we went through a series of virtual environments: islands, glaciers, deserts, even space; we manufactured incredible catastrophes; we even devoted enormous effort to construct avatars of paoxiao, hoping that the sight of others of his kind would awaken his dormant mirror neurons.
Without exception, all these tricks have failed.
Now, at midnight, only I and the zombie-like Balin remain in the lab. Everyone else has left. I know what they’re thinking: this experiment is a joke, and I’m the man who has finished telling the joke and looks around confused, unsure why everyone else is laughing.
Balin is curled up into a ball in the pet house made from pink foam boards. I remember Mr. Lu’s words. He was right. I’ve never treated Balin as a person, not even now.
A colleague once implanted a wireless receiver into a rat’s brain. By electrically stimulating the rat’s somatosensory cortex and the medial forebrain bundle, my colleague was able to induce sensations of pleasure and pain in the rat, thereby controlling where the rat moved.
There’s no qualitative difference between that and what I’m doing to Balin.
I am indeed a bastard whose mirror neurons are atrophied.
Unbidden, the memory of a children’s game resurfaces, the game in which Balin first showed us his fantastic ability.
“Shrimp! Shrimp! Watch out if you don’t want one to bite off your toes!”
I chant in a low voice, embarrassed. I pretend to be a fisherman, dipping my foot into the imaginary river from the shore and quickly pulling back.
Balin glances at me.
“Shrimp! Shrimp! Watch out if you don’t want one to bite off your toes!” My chant grows louder.
Balin stares at my clumsy movements. Gently, slowly, he crawls out of the pet house, stopping a few steps away from me.
“Shrimp! Shrimp! Watch out if you don’t want one to bite off your toes!” My legs are jerking wildly like some caricature of a pole dancer in a club.
Abruptly, Balin jumps at me with incredible speed, moving the exact way Ah Hui used to.
He remembers; he remembers everything.
Balin leaps and bounds, grabbing at my dipping leg. A baby-like gurgle emerges from his throat. He’s laughing. This is the first time I’ve ever heard him laugh in the all the years I’ve known him.
He is now re-enacting the movements of everyone in town who had been a bit different. All their movements seem to have been engraved in Balin’s brain, so vivid and precise that I can recognize who he’s replaying at a glance. In turn, he becomes the madman, the cripple, the idiot, the beggar who had broken legs and arms, and the epileptic; he is a cat, a dog, an ox, a goat, a pig, and a crude chicken; he is my drunk father and me, dancing about in joy.
In a moment, I’ve traveled through thousands of kilometers and returned to the hometown of my childhood.
Without warning, Balin begins to play two roles simultaneously, re-enacing the day of the rupture between me and my father.
Watching the argument between me and my father as an observer is eerie: the movements before me are so familiar, yet my memories have grown indistinct, unreal. I was so angry then, so stubborn, like a wild horse that refused to take the reins. My father, on the other hand, was so pitiful and meek. Again and again, he backed off; he suffered. It is nothing like how I remember the scene.
Balin quickly switches between roles, gesturing and posturing like a skilled mime.
Though I know what happens next, when it does happen I’m not prepared.
Balin wraps his arms about me, just the way my father back then wrapped his arms about him. He hugs me tightly, his head buried in the crook of my neck. I smell that familiar fish-like scent, like the sea, and a warm liquid flows down my collar like a river that has absorbed the heat of the sun.
I stay still, thinking about how to react.
Then I give up thinking, allowing my body to react and open up, hugging him bac
k the way I would hug an old friend, the way I would hug my father.
I know that I have owed this hug for far too long, to him and to my father.
I think I finally understand how to solve the problem.
At the end of The Orphans, the team that had come up with the stabbing needle test found, to their horror, that even when they harmed the aliens passing as human, their dear ones, the real humans, also failed to react. Their mirror neuron systems would not activate.
Humanity is a species that was never designed to truly empathize with another species.
Just like those aliens.
Good thing that’s just a bad piece of science fiction, isn’t it?
“We need to think about this from his perspective,” I say to Ouyang.
“His?” It takes a full three seconds for my advisor to figure out what I mean. “Who? The primitive?”
“His name is Balin. We should make him the focus, and construct an environment that will put him at ease, rather than cheap tourist scenes we imagine he’ll enjoy.”
“What are you talking about? You should be concerned about how you’re going to finish your project and get your degree, instead of worrying about the feelings of some primitive. Don’t waste my time.”
Mr. Lu once said that the progress of a civilization should be measured by its degree of empathy—whether members of the civilization are capable of thinking from the values and perspectives of others—and not some other objectified scale.
Silently, I stare at Ouyang’s face, trying to discern some trace of civilization.
The face, so carefully maintained to be wrinkle-free, is a wasteland.
I decide to work on the problem myself. Several younger students join me on their own initiative, restoring some of my faith in humanity. To be sure, most of them are motivated by their hatred of Ouyang, and it’s not a bad way to earn a few credits.
There’s a virtual reality program called iDealism, which claims to be capable of generating an environment based on brainwave patterns. In reality, all it does is select pre-existing models from a database whose brainwave signatures match the user’s—at most it adds some high-resolution transitions. We hacked it for our own use, and since our lab’s sensors are several orders of magnitude more sensitive than consumer-grade sensors, we add a lot of new measurement axes to the software and connect it to the largest open-source database, which contains demo data from virtual reality labs from around the world.