The Man Who Wasn't There: Investigations into the Strange New Science of the Self (20 page)

So, given that depersonalization makes people feel like strangers to themselves, and given that their ability to experience emotions is muted, what does this tell us about the self? “It’s telling you about the primacy of physical sensations and internal sensations” in the making of the self, Medford said. “This Damasio-type idea that feelings are built up from somatosensory information.”

Damasio himself built on ideas that date back to the late 1880s, when William James challenged existing beliefs about emotions and feelings by asking: when you see a bear, do you run because you feel fear, or do you feel fear because you run?

“
Common sense says, we lose our fortune, are sorry and weep; we meet a bear, are frightened and run; we are insulted by a rival, are angry and strike,” wrote William James in 1884, in a classic paper called “What Is an Emotion?” James suggested that this is the wrong sequence of events, however, and put forth his new hypothesis: “
We feel sorry because we cry, angry because we strike, afraid because we tremble, and not that we cry, strike, or tremble, because we are sorry, angry, or fearful, as the case may be.”

What are the modern neuroscientific definitions of emotion and feeling? An
emotion
is the physiological state of the body in response to stimuli. The state includes not just aspects like heart rate and blood pressure but also the motor movements of the body (freezing or fleeing in response to a threat, for instance). An emotion also includes the nature of cognition in that state (for example, is your thinking razor-
sharp or dull?). A
feeling
is the subjective perception of this emotional state of the brain-body complex.

When James wrote his paper, the commonsense view was that we feel first and then act, leading to the various behaviors that characterize a given emotion. So, if you saw a snake, and if you are someone who is afraid of snakes, then according to this rather intuitive view, you’d first feel the fear, and the feeling would then drive you to take action, which could be to either flee the scene or freeze in terror.

James argued that we had misunderstood the relationship between emotions and feelings: it was the other way around. And even though his use of the word “emotion” wasn’t quite in line with modern neuroscience, his larger point was well taken. We emote first and then feel the emotion.

At the same time, and independently, Danish physiologist Carl Lange proposed a near-identical idea, and so it came to be called the James-Lange theory. But Walter Cannon, who coined many words and phrases in use today in the field of physiology, including “homeostasis” and “fight or flight” to describe an organism’s response to threat, did not take kindly to the James-Lange theory. He pointed out, for example, that when people were injected with epinephrine (otherwise known as adrenaline), it produced many physiological changes that were similar to a natural emotional state of arousal, but the
people did not necessarily
feel
the artificially induced emotional state. In other words, changes in body states did not result in the expected feelings. This seemed to go against James’s notion that feelings follow emotion.

Cannon’s formidable reputation ensured that the James-Lange theory languished until the 1960s. Then, elegant experiments by Stanley Schachter and Jerome Singer resurrected the James-Lange theory, but with modifications. The scientists recruited subjects to study the
supposed effects on vision of a fictious drug called Suproxin. In reality, the subjects were injected either with epinephrine or a placebo (saline solution). When injecting, the experimenter did one of three things: (1) spoke about the exact side effects of the injection (pounding heart, shaking hands, and a warm, flushed face); (2) gave the subject a false account of possible side effects (itching, numbness of feet, headaches); or (3) said nothing at all.

Of course, the subjects all thought they had been given Suproxin. The experiment involved a further twist. When a subject was waiting for the drug to take effect, a “stooge” who supposedly had also been injected with Suproxin came and sat in the room and proceeded to vigorously act out either a state of euphoria or anger. The idea was to see if the context influenced what the subject would feel.

The experiments were revealing. The essence of the results was that feelings (of anger or euphoria) seemed to depend not just on the physiological state of the body but also on the cognitive context—which led the subject to “appraise” the emotional state of the body. The cognitive interpretation of the physiological change caused by the injection—influenced by the stooge’s behavior and what a subject had been told about the drug’s possible side effects—played a part in what the subject eventually felt or experienced. “
Cognitive factors appear to be indispensable elements in any formulation of emotion,” wrote Schachter and Singer.

A spate of follow-up experiments were somewhat inconclusive; some even failed to replicate Schachter and Singer’s results. But the general idea persisted—that feelings are the result of an appraisal of an emotional state, an appraisal that is influenced by one’s context.

Interestingly, experiments done with beta-blockers—which interfere with beta receptors all over the body and negate the effects of
epinephrine or adrenaline—were more revealing.
Beta-blockers essentially inhibit the flow of information to the central nervous system about the body’s state of arousal, resulting in reduced levels of anxiety. “
Removing cues from visceral arousal does reduce the intensity of some emotional experiences,” writes psychologist James Laird in his book
Feelings: The Perception of Self
.

Laird attributes the somewhat inconclusive results of experiments that followed Schachter and Singer’s to one factor:
the experiments did not take into account people’s differing abilities to respond to cues about the state of their own body. They did not account for varying abilities of interoception, the perception of sensations from within the body.

Still, such two-factor theories (integration of bottom-up information about the state of the body and a top-down appraisal of it) continue to be favored among emotion researchers.

Damasio and colleagues have, over the past two decades, argued that this interaction between emotion and cognition cuts both ways: while cognitive context influences our appraisal of the emotional state of the body and the consequent feelings, cognition itself can be influenced by the emotional state.

Anil Seth, co-director of the Sackler Centre for Consciousness Science at the University of Sussex, thinks that there is a better way to think about the brain basis of emotion—and take away the split between cognition and physiology. His view hews to the increasingly popular idea that the brain is a prediction machine, and that especially when it comes to external signals, what we perceive is the brain’s best guess as to the cause of those signals. Seth has extended this idea to how the brain deals with internal signals from the body and argues that it has consequences for understanding disorders like depersonalization and the bodily aspects of our sense of self.

Nicholas knows only too well the importance of feeling connected to one’s body. “I have never been so aware of what your core is as a person until I got depersonalized and felt disconnected with my physical body,” he said. “Honestly—and I’m not just saying this because I have it—I think one of the scariest things a human being can endure is the feeling of separation between your physical body and your mind, but being completely cognizant of it the whole time. It’s like being eaten alive.”

He has discussed his ongoing depersonalization with his physician. After Nicholas signed a release form, I spoke with her over the phone, and she confirmed that she had been treating Nicholas for anxiety, which had improved, but he had not shown much improvement in his depersonalization. She had referred him to a neurologist to rule out temporal-lobe epilepsy (which can sometimes cause depersonalization)—but the waiting times for specialist care in Nova Scotia being what they are, Nicholas was still dealing with his illness on his own.

I asked Nicholas about the guitar I saw in his living room. He’s learning the instrument. He’d prefer to play drums, he told me, but he is not allowed to play them in the apartment they live in. He’s waiting to move into a house where he can begin drumming again. It brings him relief from depersonalization. “Drumming requires all your attention,” said Nicholas. “When you are using all four limbs, it takes so much attention that it allows you to experience some relief.”

I was reminded of a patient Nick Medford had mentioned. The patient had been a good amateur tennis player in London but had stopped playing tennis because of depersonalization. “The only meaningful thing that I managed to do for him was to persuade him to start
playing tennis again,” Medford told me. “When he was running around the tennis court, completely immersed in the flow of that, [the depersonalization] would lift. It would come back again, unfortunately, but it was nevertheless quite significant for him, because it proved to him that it wasn’t fixed; it was malleable.”

Nicholas also pointed out that while drumming alleviated his symptoms, the relief was nevertheless transient. The moment he became aware that he was feeling better, the depersonalization returned. “It’s a paradox; you think about the fact that you are feeling better, and you start feeling depersonalized again,” he said.

Could the complex phenomenology of depersonalization be the outcome of a predictive brain gone wrong?

Think for a moment about the quandary of being a brain. It has to deduce the nature of physical reality based on constantly changing streams of sensory inputs, which are themselves modulated by the moving body. How does the brain turn stimuli into perception?

The nineteenth-century German physiologist Hermann von Helmholtz suggested that the brain solves this problem of perception by making inferences about the causes of sensations. It acts, in modern neuroscience parlance, as a Bayesian inference engine.

The word “Bayesian” comes from the Bayes theorem, which was developed by Thomas Bayes, an eighteenth-century English mathematician and clergyman.
The theorem links the conditional probability of an event P occurring given Q has occurred to the conditional probability of Q occurring given P has occurred. Bayes’s theorem is used widely in so-called Bayesian networks, which are at the heart of many modern artificial-intelligence systems. For instance, AI systems
in medicine use Bayesian networks to diagnose diseases: given a set of symptoms and test results, the system calculates the probabilities for various causes, offering up as a diagnosis the one with the highest likelihood of causing the symptoms. As another example, consider the case where a patient tests positive for Ebola, but the test itself is accurate only nine out of ten times. What is the probability that the person has Ebola? Our naïve intuition will want to rely on just the test and say the probability is 0.9. Our intuition would be wrong. The probability depends where the person is being tested. If he or she is in a country where Ebola is endemic, the conditional probability that the person is infected is higher than if the person is tested in a country where there is zero incidence of the disease.