‘Many fairy tales about the brain still propagate through our field’
Our editor Jon Sutton poses the questions on the eve of the publication of Lisa Feldman Barrett’s new book, How Emotions Are Made.
31 January 2017
Your new book is called ‘How Emotions Are Made’. Can you explain that word ‘made’?
For much of recorded history, scholars have been battling over two different views of emotions. One view, the classical view of emotion, proposes that happiness, sadness, fear, and several other emotions are built into your brain and body from birth, i.e. they are natural kind categories. Scientists who take this view ask questions like, ‘Which brain region contains the circuit for fear?’ The question assumes that fear lives in a specific set of neurons within the brain. They also ask, ‘What is the facial expression for fear?’ and ‘What is the bodily state of fear?’ These questions assume that fear has a physical fingerprint (a pattern in the face and body) that is specific to fear, and that this fingerprint can objectively identify when fear has been triggered.
The other approach, what I call the constructionist view of emotion, proposes that an emotion like fear is not a natural kind category, but is, instead, a population of highly variable instances. For example, you can tremble in fear, jump in fear, freeze in fear, scream in fear, gasp in fear, hide in fear, attack in fear, and even laugh in the face of fear. In each case, your face and body might take on a different pattern. Each instance of fear (or any other emotion category) is created in the moment, for a specific situation, from more basic ingredients – core networks in the brain that are wired, in part by the kind of body we have and the social surroundings we live in. Scientists who take this view ask more contextual questions, such as ‘What does fear look like in this situation, when a person is trying to achieve this particular goal?’ They also ask computational questions, such as ‘How does the brain create this specific experience of fear?’ without presupposing a single implementation of fear in the brain such as a dedicated circuit.
Which argument is winning out?
The scientific evidence that supports the classical view of emotion sits within a much larger body of evidence that disconfirms it. In the last 10 years, neuroscience evidence has pretty definitively handed the victory to construction, thanks to the discovery that the brain is a predictive organ, not a reactive one (called predictive coding) and that psychological phenomena cannot be localised to particular blobs in the brain, but are instead built with intrinsic brain networks that are multi-purpose. Emotions are made, not built-in. However, the news hasn’t yet reached the public: hence the need for my book.
What specifically have we – as psychologists and as everyday students of our selves – been getting wrong?
Many fairy tales about the brain still propagate through our field. Here’s one myth: the brain contains circuits that sit in the ‘off’ position until something in the world flips them to the ‘on’ position. In reality, your neurons fire all the time – this is known as intrinsic brain activity – and sensory input from the world only briefly perturbs the patterns of activity, like spitting into an ocean wave.
Here’s another myth: that cognition and emotion are separate in the brain, and the former regulates the latter. Back in the 1950s, Paul MacLean proposed a model of the brain with a reptilian core for appetites, such as hunger and sex, cloaked in a mammalian ‘limbic system’ for emotion, which itself is controlled by a cerebral cortex for rationality. This model, called the triune brain, is a fantasy. It’s Plato’s allegory of the charioteer and his two horses, tattooed on the brain. But the brain didn’t evolve in layers like sedimentary rocks. Rather (in the words of the neuroscientist Georg Striedter), brains evolve like companies do: they reorganise as they expand. The brain regions that MacLean considered emotional, which he referred to as the ‘limbic system’, are now known to contain major hubs for general communication throughout the brain. They control the various systems of the body, and they’re important for many phenomena besides emotion, such as including language, concepts, stress, and even the coordination of the five senses into a cohesive experience.
And here’s a new myth, based on a very old idea: pattern classification, a technique from artificial intelligence, has supposedly discovered the neural essence for each emotion. Scientists who make this mistake examine many brain scans for, say, fear, and compute a statistical summary. But then they point to the summary and say, ‘Look, the essence of fear in the brain!’ That’s a gigantic logical error. They’re mistaking a mathematical average for the norm.
Some other popular misconceptions are that Charles Darwin allegedly wrote that facial expressions are functional products of natural selection, and that William James allegedly described how every type of emotion has a distinct bodily state. These ideas are taught as facts in universities, but neither scientist made these statements. I provide the full story in my book.
How do others respond to your view of emotion? Do you see it is a positive and liberating one?
Engineers and computational neuroscientists find my view intuitive. Many psychologists respond initially with disbelief, because their experiences guide their beliefs about how nature works. Emotions feel given, and so it seems common sense that marvelous experiences like surprise and gratitude would be physically distinct, built-in, and even universal. But our personal experiences do not reveal how nature works, any more than our experience of watching the sun cross the sky means that it revolves around the Earth. The brain is a master of sleight-of-hand. To believe otherwise is naïve realism.
Once people understand my theory of constructed emotion, they find it positive and liberating. You have much more control over your emotional life than the classical view of emotion would imply, with its stories of emotion circuits that trigger automatically and are regulated after the fact. If emotions are constructed from more basic ingredients, you can alter the ingredients and change your life. For example, you can make yourself more emotionally intelligent by learning more emotion words, giving your brain new concepts to predict with, so you can deal more flexibly with a broader range of situations. That kind of change is within your control. I want to be clear that control does not mean overcoming a reflex reaction in the moment. By cultivating experiences to curate your concepts, including emotion concepts, you can tweak your brain’s automatic control of your future experiences and behaviors. In a literal, brain-wiring sense, the concepts you learn today influence your life tomorrow.
Can you give me some other practical implications of this view of emotion?
Suppose you’re in a doctor’s office, complaining of pain in your chest. If you’re a woman, you’re more likely to be diagnosed with anxiety and sent home, whereas if you’re a man, your symptoms are more likely to be understood as the first signs of a heart attack and you’ll receive lifesaving preventive treatment. As a result, women over age sixty-five die more frequently of heart attacks than men do, because the perceptions of health care professionals are shaped (wrongly) by classical view beliefs that they can objectively detect emotions like anxiety and that women are more emotional than men.
Another example: right now, companies are spending billions of dollars building devices and apps to objectively detect emotion, on the mistaken belief that we can read other people’s emotions simply observing their facial movements, heart rate, and so on. These ideas are the classical view’s bread and butter, but they are not supported by the majority of scientific evidence. The stereotyped facial expressions for emotion (e.g. scowls for anger, pouts for sadness) that are used in experiments were stipulated by scientists, not discovered by them. No study has ever found the elusive bodily state that is unique for each emotion (i.e. that is consistently specific to that emotion) despite the fact that hundreds and hundreds of studies have been searching for the last century. Once you become aware of the full range of scientific evidence, it becomes clear how much of the money and effort that fuel the emotion economy is wasted while other, more viable scientific questions are ignored. Emotions can’t be found in facial movements and heartbeats alone. They are not simple body signals. Emotions are far more complex constructions that involve a person’s past experience. The theory of constructed emotion in my book provides a more evidence-based approach to build machines that perceive emotion.
To find emotion’s ‘fingerprints’, do psychologists need to focus on different methods?
It’s always a good idea to use multiple methods in any scientific study. But using a menu of methods won’t suddenly reveal support for the classical view. An emotion, like anger, is not a thing. It is a category of events. Any two instances of, say, anger may be constructed by entirely different sets of neurons and be associated with entirely different bodily states and facial movements. Variation is the norm. We know this from meta-analyses of hundreds of studies and tens of thousands of test subjects. The theory of constructed emotion explains why this is the case, and how highly variable emotions are made.
If psychologists want more bang for their buck, they should start asking different questions. For starters, they should stop asking questions about emotion that begin with ‘where’, such as where is happiness located in humans, where is the fear circuit in a rat, and instead, ask ‘how’. How are instances of happiness and fear created?
I note in the book’s acknowledgements you say you wrote it for your daughter, ‘to understand the power of your own mind… You are an architect of your experience, even in times when you feel buffeted by the world.’ That seems like a pretty empowering message, but I guess some people would just find it overpowering?
The theory of constructed emotion, when compared to the classical view, is an ambassador for a radically different view of what it means to be a human being. Part of that view, which I lay out in the book, goes like this: We are not victims of primitive, animalistic emotion circuits. Your brain actively constructs your emotional experiences, even those that feel out-of-control. If you understand the construction process, you can learn to influence it. I guess it’s possible that some people may find this message overpowering, but to me, it’s freeing.
Talking of the acknowledgements, that’s got to be one of the longest acknowledgements sections I’ve seen. Does that suggest you’re someone who wears their heart on their sleeve?
Every sole-authored paper in science is an illusion — no scientist works alone. Every scientific discovery is made in the context of other working scientists. I believe it’s important to give credit where credit is due. I have been fortunate to have colleagues, students, research program officers, and so on, who have created a context for me to write How Emotions Are Made, and it is a pleasure to be able to offer my thanks in a formal way. Also keep in mind that the research behind my book comes from multiple disciplines. I began my career as a clinical psychologist but have since gained expertise in social and personality psychology, physiology, cognitive science, neuroanatomy, neuroscience, evolutionary biology, and more recently in computational methods. Each transition was made possible by generous colleagues (and funding agencies). The theory of constructed emotion could never have taken shape without their support.
But would you say you are somebody who enjoys experiencing the extremes of emotion?
Well, my family, in conjunction with my lab, runs a haunted house for charity every Halloween, donating all the proceeds to our local food bank. My colleagues, students and staff become wandering monsters in the haunted house, and we use our research knowledge of fear to scare the daylights out of kids and adults. We never use gore or cheap thrills, just careful positioning, ambiguity, and misdirection. It is a team-building exercise for the lab. Everyone has a lot of fun, ghouls and guests alike. I enjoy creating contexts where the people I care about come together in community.
Who are your inspirations in the psychology community, the roots of your approach?
Historically, I’d say William James, Wilhelm Wundt, Hermann von Helmholtz, and, of course, Charles Darwin (for On the Origin of Species, not The Expression of the Emotions in Man and Animals).
As for the modern day, I am inspired by so many people who appear in the acknowledgments of How Emotions Are Made. Jim Russell for his mentorship, for his concept of core affect, and for creating the affective circumplex, which is so ubiquitous now that people no longer cite him for it. Larry Barsalou for his theory of grounded concepts that solidified my theory of constructed emotion. Barb Finlay for knowing, basically, everything about everything. For a more complete list, see the book.
You say your husband often asks ‘Is this for the 1 per cent?’, as in your scientific colleagues as opposed to a general audience. Any tips on how to handle that balance?
When explaining how science works, or what scientific studies have revealed, my natural inclination is to include all sorts of details. That’s what we are trained to do, as scientists. Often, I think the details are fascinating. Sometimes, I’m defending a counterintuitive idea that people find provocative, and so I want to offer a slew of evidence. But good writing is knowing what to leave out.
For example, much of the neuroscience in my book concerns visceromotor circuitry in the brain, which is largely responsible for regulating your body’s energy needs via the autonomic nervous system, the immune system, and the endocrine system. ‘Visceromotor’ is a mouthful, however. This circuitry is also called ‘limbic’, a shorter name but loaded with historical baggage. After much deliberation (and some agony), I discarded both names and opted for something simpler. The body has a multifaceted energy budget, much like a financial budget, and visceromotor regions of the brain operate kind of like a company’s financial division by overseeing that budget. So I invented the phrase ‘body-budgeting regions’ to explain the functions of these brain regions more intuitively.
Perhaps the best tip I have is: aim for the number ‘1’. Each paragraph should contain just one idea. Each phenomenon should be called by only one name (even if scientists have 12 different names for it). Each talk should have one punchline. ‘1’ is still a struggle for me. In my academic writing and speaking, I still end up closer to ‘2.5’.
That’s why it’s helpful to have someone else to keep you honest. For me, that was my husband, first and foremost. But I was also fortunate to have several close friends who were dedicated readers, plus my editors at Houghton Mifflin Harcourt and the New York Times. If you decide to write about science for the public, make sure to get feedback from plenty of non-academic test readers, and edit, edit, edit. When you believe you are finally done, edit one more time. I wrote three complete drafts of How Emotions are Made before settling on the final narrative. There is no shortcut to effective writing.
What does the future hold?
I remain hopefully that my future contains a beach vacation where I can sleep late every morning and read novels all day. But when it comes to the science of emotion, I hope that in the coming years, we’ll all see fewer and fewer news stories about brain blobs for emotion in people or rats or fruit flies, and more about how brains and bodies construct emotion.
More broadly, I am confident that neuroscience will reveal the brain to operate even more via construction than we now know it does. I hope this progress continues to promote a full scale scientific revolution in psychology. I’d like us to finally trade in psychology’s 17th century theory of mental faculties (where every phenomenon has its own process), its 19th century stimulus-response approach to experimentation (which forces the brain into an unnatural state that is rarely encountered in the real world), and its 20th century reductionist search for universal laws. We will create a more powerful, robust science if we better appreciate that experiences and behaviours derive from populations of time-varying, context-dependent brain states. This, along with a new philosophy of psychology that is modeled on modern biological concepts of population thinking, emergence and complexity, will go a long way toward solving psychology’s so-called ‘replication crisis’.
- How Emotions Are Made: The Secret Life of the Brain is out now.