The Man Who Wasn't There: Investigations into the Strange New Science of the Self (21 page)

A Bayesian brain, in theory, functions similarly. It computes the most likely cause of sensory inputs based on prior beliefs it has about the probable causes of such inputs. As alluded to earlier, the brain’s best guess as to the cause of the sensations rises up, so to speak, as perception. Of course, this is an ongoing process. The brain uses internal models of the body and the world to predict the expected sensory input. Any difference between the expected signals and the actual signals constitutes a “prediction error.” The brain uses these error signals to update its bank of prior beliefs, so that it can predict (and thus perceive) more accurately when similar signals come again.

Such “predictive coding” models, which use Bayesian inference, have mainly been applied to explain exteroception: making sense of external sensations coming from outside the body. Interoception also involves perception, but in this case it’s making sense of the signals from within the body. The brain needs to know the body’s state in order to determine if the body has moved away from its biochemical comfort zone and whether it needs to initiate actions to get the body
back to a physiological state that’s optimal for survival. Anil Seth’s argument is that predictive coding should hold true for making sense of internal bodily signals too. “It is just another process of perception,” he told me.

Seth’s argument has a bearing on emotions and feelings. The two-factor models of emotion always involve the integration of information coming into the brain via incoming nerves, generating a snapshot of the physiological state of the body, which then is subject to cognitive interpretation and subsequently gives rise to the feeling of an emotion. With predictive coding, the division between cognition and physiology is done away with. At no place in the brain is there incoming information being integrated to create a perception. Rather, it’s prediction all the way—what you perceive and what you
feel
is always the brain’s guess as to the cause of the signals. This sophisticated idea evolved from the same thinking that gave us corollary discharges and comparators, models that have been used to explain how the brain gives rise to a sense of agency and suggest their impairment may lead to the symptoms of schizophrenia. Turns out that predictive coding can be applied to just about anything and everything the brain does.

Seth argues that there would be many levels of predictive coding in the brain. The lowest level would be predicting the causes of the incoming sensory signals from the body. The prediction would form an input, or incoming signal, to the next level in the brain and so on. It’s a hierarchical model that is well suited to how the brain is structured. “The subjective emotion we feel is the brain’s best predictive guess that explains the [incoming] interoceptive information at a whole bunch of hierarchical levels,” said Seth. “It’s not just cognition looking down at physiology and interpreting it.”

Predictive coding is a new way of thinking about the brain and
how it carries out its remit, with the potential to explain interoception and exteroception, emotions and feelings, and indeed explain how psychopathologies emerge when this predictive mechanism goes wrong. As we’ll see in the next chapter, the model is even being applied to explain something as complex and varied in its symptoms as autism. But there is “also a danger,” said Seth: “You explain everything, you explain nothing.”

The main criticism of the predictive-processing approach is that there is no direct evidence for it. But there is evidence consistent with predictive coding. For instance, the work on the mechanisms of corollary discharges/comparators is considered circumstantial evidence for predictive brain mechanisms. There’s also evidence that the insular cortex, tucked beneath the temporal and frontal lobes, is the brain region most likely involved in doing comparisons between top-down predictions of expected interoceptive signals and incoming signals that contain information about prediction errors.

Accepting such circumstantial evidence for now, the big question for psychopathology is what happens when there are errors in prediction. Errors are an indication that whatever the brain is modeling at any given level is not quite right. There are two options at this point: the brain can either update its model and bring it in line with the sensory inputs; or it can initiate some action to push the body to the desired state. The latter mechanism would be the basis for homeostasis (say you wade into the frigid waters off San Francisco and stay in a little too long—your internal body temperature will start dropping below limits that are deemed acceptable by the brain’s model of your viscera, and you will feel the urge to get back to warmer environs).

In this way of thinking, the brain’s function is to minimize prediction errors. And this has consequences for our sense of self. Take
the signals that are coming from one’s own body. In Seth’s thinking,
when the brain’s internal models are accurate, and there is a good match between the predicted and actual interoceptive signals, you get a sense that you are embodied, that your body and emotions belong to you. The match between predicted and actual signals implicitly tags the body and associated emotions as “self,” whereas a mismatch would be akin to tagging it “non-self.” So, the vividness of emotional feeling states and the sense that they are
mine
depend on the brain making accurate interoceptive predictions and minimizing the corresponding prediction errors.

But what if there are persistent and ongoing prediction errors, either because of faulty internal models of the body in the brain or because something goes wrong in the neural circuitry that compares and generates errors (it’s intriguing that the insula, which is key to models of predictive coding and interoceptive inference, is one of the brain regions implicated in depersonalization)?

Seth speculates—and stresses that it’s just speculation—that this would then lead to dissociation, a sense of unreality about one’s own body and emotions, feeling disembodied, feeling estranged from oneself. It’s as if the predictive brain’s best hypothesis for the source of interoceptive signals—given persistent errors—is that the signals do not belong to self, but rather to non-self.

As with everything we have encountered so far, depersonalization too does not destroy the “I.” There is still the subjectivity—the self-as-subject—that is aware of being estranged from other aspects of the self, in this case the vivid emotions and feelings that give us our sense of being embodied. So, while no one is saying that our emotions and feelings are not integral to our sense of self, it’s nonetheless intriguing
from a philosophical perspective that they don’t
constitute
the self-as-subject; the “I” stands apart, watching, observing.

I watch Nicholas with his thirteen-month-old daughter. He cuddles her in his arms, kisses her, almost as if he’s not going to let her endure the lost childhood that he did.

Is he in touch with his father? I ask him.

“I don’t talk to my father anymore,” he says. “Stopped talking to him about a year ago.”

“Anything trigger that?” I ask.

He hesitates.

“You don’t have to talk about it if it makes you uncomfortable,” I say.

But Nicholas does talk. When he lived with his father nearly a decade ago, his father had been a muscular, tattooed man, “very hard-looking in the face . . . like a criminal.” But then he went west, to Alberta, and got into harder drugs. By the time he came back to Nova Scotia, he had dropped seventy pounds and was now a small-framed man, with a weathered face and an alcoholic’s nose scarred with broken blood vessels, “an extremely, extremely messed-up person.”

In October 2011, things got worse. The police in Bridgewater received a 911 call from Nicholas’s father. They arrived to find one of Nicholas’s close friends, a young man of twenty-two, dead in Nicholas’s father’s apartment. (It was later determined that the young man died from a methadone overdose. Nicholas’s father would be charged with prescription drug trafficking, but the case was dropped for lack of evidence.) Soon after that tragedy, Nicholas’s father got into a relationship with the young man’s girlfriend, who was just eighteen.

“You can almost hear Jerry Springer in the background,” quips Nicholas.

The saga had an even more sordid end. In March 2012, the young woman was found unresponsive in an apartment and she too died, in a hospital in Halifax two days later—allegedly due to a prescription drug overdose.

Nicholas’s father did end up in jail eventually, for drunk driving.

Nicholas’s mother suffers from auditory hallucinations and paranoid delusions (possibly the result of years and years of drug abuse). She was blond, thin, and pretty when she was young. Now she’s disproportionately overweight, possibly due to her medication, with extra pounds on her face and midriff, and looks older than her age, says Nicholas. He meets her occasionally. “It’s not much of a mother-son relationship,” he says. “We never really had that.”

Does he feel sad it’s turned out this way with his parents? I ask him.

“I do, definitely,” he says. “At the same time, with depersonalization, I feel disconnected from it. It’s odd—I feel sad about it, [but] it doesn’t feel like they are my emotions. It almost feels like I’m feeling sad about somebody else’s life story.”

“You have been through a lot,” I say.

“Yes, I have,” says Nicholas. “It’s been one hell of a life. I wish it felt more like my life, and not like I’m observing somebody else’s life.”

“Even though it’s been so harsh?”

“Oh, yeah,” he says. “I wish I could come to terms with it. It’s hard to come to terms with something when it doesn’t feel like it fully happened to you.”

We talk of this and that. Of the big lizard—an Australian bearded dragon—that he keeps in an aquarium. Of San Francisco and the
Golden Gate Bridge, from where I have come. Nicholas and his fiancée are enamored by the idea of California. I point out they are living in a spectacularly beautiful place themselves. “You can’t appreciate it as much if you have been here your whole life,” says Jasmine.

She could be speaking about depersonalization and the estranged self. Those of us who inhabit our bodies seamlessly and without disruption and feel and own our vivid emotions may not value what we have. You can’t appreciate the self as much if you have been intimately connected to it your whole
life.

THE SELF’S BABY STEPS

WHAT AUTISM TELLS US ABOUT THE DEVELOPING SELF

Autists are the ultimate square pegs, and the problem with pounding a square peg into a round hole is not that the hammering is hard work. It’s that you’re destroying the peg.

—Paul Collins,
Not Even Wrong: A Father’s Journey into the Lost History of Autism

I myself am opaque, for some reason. Their eyes cannot see me. Yes, that’s it: The world is autistic with respect to me.

—Anne Nesbet,
The Cabinet of Earths

J
ames Fahey was thirty-four when he was diagnosed as being an Asperger. When I referred to his “Asperger’s syndrome” in our first correspondence, James, without taking any offense, gently chided me: “
I am an Asperger
, pure and simple—I do not have or suffer from any artificially constructed syndrome, disorder, disease, or flaw.”

His diagnosis came over two sessions with a psychologist. During the second session, his younger sister joined him, to provide the psychologist some perspective on their childhoods together. It was then that James realized for the very first time that some of his perceptions of those early years differed from his sister’s. It came as a shock.

For instance, his sister recalled clearly that James had never wanted to play with her, even if she was alone and he was by himself too. She wanted company, and he didn’t recognize that. “I never thought that she might get lonely,” said James. “I don’t get lonely; why would my sister get lonely? It just never occurred to me.” James would rather read by himself; even his choice of books (usually historical accounts of wars—the Napoleonic wars, for instance) set him apart from his sister. He was interested in facts and statistics. She would read children’s fiction. “I can’t stand fiction,” James told me. “I can’t see the reason why I would ever want to read it.”

James grew up in the western suburbs of Melbourne, Australia. His parents weren’t particularly physical in their affection, which was fine by him. “I won’t say it was a cold or neglectful upbringing; my sister came out fine,” he said. James was uncomfortable being touched or hugged. His extended family on his mother’s side didn’t pose problems: they were mostly farmers, living out in the countryside. “They stood away from you; they didn’t need to have a close-proximity conversation,” recalled James. “They didn’t want to hug you all the time.”

Interactions with his father’s family, however, were an ordeal. Most Sundays, James’s family would drive over to his paternal grandmother’s place, where she’d be waiting, at times with his great-aunts. They’d want to hug and kiss him. “At the time, I couldn’t think of anything worse than someone planting a big, sloppy, slimy kiss on
me,” said James. “And I didn’t like being hugged. I felt trapped, locked in a cage.”

He was well into his twenties when he began realizing that who he was and people’s notions of how he should live his life were incompatible. Having a girlfriend, for instance. Social pressure made him try, but he found it difficult. “How many girls are going to accept a guy who wants to go and be by himself all week?” he told me. “That’s not much of a relationship.” And the fact that he didn’t like touching didn’t help matters either. Any intimacy had to be structured. “Want some intimate moments, that’s fine, if it’s organized at a particular point in time. Go to bed, you do your thing, that’s fine. Don’t touch me again.”

It’s not that James is against relationships, he just doesn’t want the intimate kind. “I’m not sad or depressed about this, not in the slightest, given how much I enjoy and need solitude. Platonic relationships suit me fine. They provide most of the positives, without most of the negatives.”

Even in less-demanding social interactions, James is aware that he can get things wrong. “Maybe laughing inappropriately, schadenfreude moments,” he said. Once, while watching a movie with friends, James began giggling about something he noticed in the background. It had nothing to do with the main story line. But his mirth drew attention. The movie was
Schindler’s List
. His friends thought he was amused by Jews being rounded up. But James had been fixated on something minuscule, unrelated. In a way, it’s how he sees the world around him. To explain, he referred to
Schindler’s List
again—there’s a scene in the mostly black-and-white movie in which the main character, Oskar Schindler, sees a girl in a red coat. Later, we see a red coat among a pile of bodies being carted away. Our attention is drawn to that splash of color. “That’s kind of a fair analogy for how I see things
in the world,” James said. “Some things are highlighted to me. I follow them around. I don’t know why I assign this degree of interest to particular things.”

Well before he got his official diagnosis (a term he hates—“makes me sound like I have a problem”—even though he values much of what psychiatrists do), James had thought long and hard about why he was different. Socializing made him anxious, depressed. Even today he’s anxious around most people. His introspective nature, however, helped him make sense of who he is and eventually shed the burden of others’ expectations. “I mistook my own self for what they believed I should do,” he said. “They just couldn’t understand my perspective.”

The kind of traits that James’s introspective insight laid bare—his desire to be alone, his acknowledged difficulties in forming intimate social bonds, anxiety in social situations, and traits from his childhood, such as an aversion to being hugged and kissed—came under scrutiny in the early 1940s, when the word “autism” was used for the very first time in medical literature.

“Autism” comes from the Greek word
autos
, meaning “self.”
In 1916, Swiss psychiatrist Paul Eugen Bleuler coined the term to describe a symptom of schizophrenia—a symptom that has been described as “
the narrowing of relationships to people and to the outside world, a narrowing so extreme that it seemed to exclude everything except the person’s own self.”

Then, in 1943, Leo Kanner, an Austrian-born psychiatrist and director of the Children’s Psychiatric Service at Johns Hopkins Hospital in Baltimore, wrote a landmark paper in which he used the word
“autism” to describe the syndrome with which it’s now associated. It was an elaborate paper, detailing keen observations of eleven children:

The . . . fundamental disorder is the children’s
inability to relate themselves
in the ordinary way to people and situations from the beginning of life. Their parents referred to them as having always been “self-sufficient”; “like in a shell”; “happiest when left alone”; “acting as if people weren’t there”; “perfectly oblivious to everything about him”; “giving the impression of silent wisdom”; “failing to develop the usual amount of social awareness”; “acting almost as if hypnotized.”

Kanner differentiated his notion of autism from Bleuler’s account of the symptoms of schizophrenia, in which the withdrawal from normal social relationships begins late in childhood, or even in adulthood. About this new syndrome, Kanner wrote, “
There is from the start an
extreme autistic aloneness
that, whenever possible, disregards, ignores, shuts out anything that comes to the child from the outside.”

In a notable illustration of the emergence of an idea whose time had come, Hans Asperger, an Austrian pediatrician living in Vienna, independently published a similar paper of case studies a year later, also using the word “autism” to characterize the condition.

It wasn’t until 1980, however, that the
Diagnostics and Statistical Manual of Mental Disorders
(DSM), the oft-quoted, much-maligned publication of the American Psychiatric Association, first included autism as a diagnostic category, albeit with the unfortunate phrasing: “infantile autism.” This was renamed in 1987 as autistic disorder. But the difficulty in defining autism became evident in the DSM-IV (1994): autistic disorder was now placed alongside several more subtypes,
including Asperger’s disorder and pervasive developmental disorder not otherwise specified (PDD-NOS). In 2013, the fifth edition of the manual (DSM-5) reversed course, rolling back Asperger’s and PDD-NOS into the umbrella term
autism spectrum disorder
.

Through all this categorization and recategorization, Kanner’s original insight remains: “
We must, then, assume that these children have come into the world with
innate
inability
to form the usual, biologically provided affective contact with people, just as other children come into the world with innate physical and intellectual handicaps,” he wrote in 1943. Or, as he put it, the children had “innate autistic disturbances of affective contact,” where “affective” refers to the emotional aspect of our being.

The function of a self, by definition, is to help an organism discern the boundary between itself and others. Is an infant born with this ability? Or does a developing baby form a self in stages? There’s certainly good evidence that by about eighteen to twenty-one months of age, babies begin to distinguish between themselves and others while speaking. To the horror of some parents, infants start screaming “mine” for things they fancy, a clear indication of the ability to
explicitly
reference one’s self using language. Around the same time, a child is able to recognize his or her reflection in a mirror or in photographs. One of the questions facing developmental psychologists is this: does the child develop a sense of self primarily through social interactions and the use of language, or is there an underlying, more innate, and implicit self?

William James distinguished between the implicit and the explicit, calling the former “I” and the latter “Me.” As psychologist Philippe Rochat puts it, “
The ‘Me’ corresponds to the self that is identified,
recalled, and talked about. It is the conceptual self that emerges with language and which entails explicit recognition or representation. It is beyond the grasp of infants, who by definition are preverbal, not yet expressing themselves within the conventions of a shared symbol system. On the other hand, there is the self that is basically implicit, not depending on any conscious identification or recognition.”

In 1991, Ulric Neisser, often called the father of cognitive psychology, further split James’s “I” into an
ecological self
(an implicit sense of the body that babies develop in relation to their physical environment) and an
interpersonal self
(an implicit sense of the social self that emerges via interactions with other people).

The ecological self is evident in babies. For example, babies have a rooting reflex: touch them on the cheek and they will turn their heads toward the touch.
Rochat has shown that even newborns, within twenty-four hours of birth, are three times more likely to show rooting behavior when someone else touches their cheek than when they accidently touch their own cheek. This shows that babies are aware of their own bodies, maybe even as agents of their own actions (recall from the chapter on schizophrenia that we are unable to tickle ourselves because the efference copy/corollary discharge mechanism dampens down the sensations; we implicitly know we initiated the tickling).

According to Rochat, babies develop an implicit social self because adults are constantly mirroring the facial expressions and emotions of babies (how often have you seen a mother sympathize with a crying infant by making a sad face herself and saying, “My poor baby”?). Babies in turn are themselves imitation machines. It’s this constant mirroring that helps the baby develop a preverbal, prelinguistic social self—one that is honed by interpersonal interactions.

With language comes the ability to form and express the explicit
self. As they grow further, children develop another intriguing ability: the ability to peer inside the minds of others, so to speak. In children with autism, however, this ability is impaired, leading to what Kanner called an innate inability to relate socially to others. For researchers who are also thinking of autism in the context of the self, questions loom: is the problem of social-relatedness linked to the child’s own developing self and, if so, how?