Main pageWritingSourcesRecipesDoodles

How Emotions are Made

How Emotions are Made: The Secret Life of the Brain by Lisa Feldman Barrett (henceforth HEAM and LFB). This is a book that really changed how I perceive my experience of the world, and some of the concepts from it really seem to support concepts from other books that have made a big impression on me.

Personal/Historical Notes

So, I picked this book up about six years ago, shortly before my first kid was born (I think – pretty sure I brought it to the hospital with us, just in case I had some downtime). Then, about five chapters in, I suddenly had an infant to take care of and discovered I had very little time or attention for reading. Recently I’ve been bookclubbing this, and enjoying it just as much, and seeing the ways that reading what I did years ago has affected my thinking since. I’m actually only about halfway through, so we’ll see how the rest goes, but I think what’s been covered so far is pretty important so I’ll try to summarize the important bits for linking to elsewhere on the site.

One more note about reading this: Back in freshman year of college I was pretty into psychology, neuroscience, and the idea that my brain was some prehistoric creation messing with my ability to be a perfectly logical thought-being (I blame The Science Times and The Primal Teen). This was before I had read much feminist literature (it must have been first semester freshman year, because Woman and Nature was part of Applied Ethics second semester), so I was still deeply suspicious of emotions and seeking to understand and distance myself from them. I found some references to this reasearcher Paul Ekman who was researching emotion and how it shows up on people’s faces. He – along with some collaborator whose name I don’t remember name, though it might have been his mentor: Silvan Tomkins – wanted to take pictures of all the different expressions the human face can make, but discovered there were muscles they didn’t seem able to control so they stuck electrodes into their faces to trigger the muscle movement. This struck me as pretty Punk, and his book Emotions Revealed (pdf link) had just come out, so I guess I noted down that I wanted to get it? My memory is a little foggy with respect to when I bought/read it, though I remember reading it in Art House, so sophomore year at the earliest (ah! my email history informs me I bought it spring semester sophomore year).

Anyway, Emotions Revealed was a big component in my starting to accept emotions. It defines a set of emotions and claims that they’re universal: all humans across cultures are supposed to experience these emotions. It then provides a nice description of what you might feel as you’re experiencing each emotion, and talks a little about things that are emotion-related like “moods” (which can last much longer than an emotion) and what Ekman calls “scripts” (when you react emotionally to a current situation based on some much older situation, that’s a script). LFB spends the first part of HEAM just tearing Ekman and his work apart. I mean, the broad strokes still kinda hold up: “moods” and “scripts” are still referring to something real (as real as money or more so), but universality of emotion/emotion faces/etc. all get dismantled. I don’t think this book would’ve made as big an impression on me if it didn’t blow up the foundations on which my previous understanding of emotion had been built.

Interesting concepts

These aren’t necessarily the first place I’ve encountered these concepts. Also, I might be smooshing some of these together, or calling them by slightly the wrong name.

Population Thinking

There is no one dog in the world that is the archetype of a dog. Each dog (each person, each flower, each grain of sand) is unique, with no two the same. They are a population.

LFB says our emotions are also a population: no two instances of us feeling an emotion are precisely the same, nor are two different people’s experiences of an emotion the same.


A concept is an abstract thing existing in our minds that groups things together. We don’t have a list in our heads of all the things that match the concept, we just kinda check things on the fly. Your and my concepts might not quite match, either.

Take “car” for example. We all know what a car is, right? What if it’s got three wheels? two? Treads instead of wheels? See also: is a hotdog a sandwich?

The Brain as Prediction Engine

This is one of the bits that I basically already understood from reading other neuroscience/psychology stuff (for a good, short article, see here). Our brains spend all their time predicting what’s about to happen, rather than just reacting to what the senses say has happened. When you think you see someone on the street but then it’s not them? Prediction Error.

One of the oft-trotted out facts in support of this is that our brains have many more pathways “down” from our “higher-level” brain regions to our sensory regions than going “up” from our sensory regions. When I was a teenager (see above) I understood this as our brain being able to “override” what our senses say. LFB presents it more as a negotiation, with our brain constantly guessing and being “held in check” by our sensory input. My sense is that those “down” pathways are probably how most of us can see pictures in our heads, or imagine what something will taste like.

A key part of this concept for me is that what has happened recently is always feeding into what we predict is going to happen next, which I think explains the “priming” that so many psychology studies depend on. “Make ’em read a happy or sad paragraph and then play this game.”

Interoception/Energy Budgeting

Much of what we call emotions, LFB says, are based on interoceptive experiences from our energy budgeting systems. Interocpetion is a sense (like taste or sight) of things going on in your body: hunger, need for oxygen, fatigue. Our energy budgeting systems have evolved to help us to conserve energy or use it efficiently. She uses as an example one of her Ph.D. advisees wearing a heart rate monitor, and it beeping loudly when the student saw her, implying that the student’s body was preparing her for some sort of fight-or-flight or other stress response.

This is the part that, I think, most rocked my world, because it explains so many other things. We have these bodily sensations (I’ll call them “feelings”) and we ascribe “emotional meaning” to them, but the’re not inherently emotional.

And paired with the Prediction Engine stuff above I think this explains why we have what Ekman calls “Scripts”. Certain situations “prime” us to react in certain ways. We learn what emotions mean for us, and develop habits of how we react. If emotions are mostly built-in, as Ekman lays out in Emotions Revealed, then scripts are kind of a weird thing. If emotions are purely learned, then “scripts” are just… how we react to things. Our brains/bodies learn that when we’re in a situation that feels like this and we react that way we get an energy reward, so when we’re in similar-feeling situations in the future we should react the same way.

I feel like there’s also a lot to say about trauma here, especially capital-t Trauma, but I’m probably not the person to say it.

“As real as money”

I don’t think she actually uses this exact phrase in the book, but she uses some similar phrases. I love it. I want to use this all the time to talk about things people think of as “real”.

Some Criticism


Sometimes LFB comes up with definitions that… just don’t work for me.

At the beginning of Chapter 7, “Emotions as Social Reality”, she goes into how if a tree falls in the forest with no one around to hear it doesn’t make a “sound” it just makes waves of pressure in the air. “Sound” is an observer-dependent interpretation of those pressure waves, she says. And like, okay, if you need to set up this distinction to make your point that’s fine, I guess. It’s hard to talk about stuff when we don’t have good words for it, and sometimes you need to make some, but she’s saying it as though this is the one way to define sound, right before spending pages and pages talking about how most things are subjective and I dunno… I just wish it were done better.

Then we get real deep into semantics when she’s talking about how we can’t experience an emotion without a concept for it, and sets up an analogy with looking at a rose if you have no concept for flower. She says “No scientist would claim that you’re seeing a flower but just ‘don’t realize it’”, but… what? Why not?

I think I could find a set of definitions for experience, emotion, definition, knowledge, etc. that would make her sentence hang together, but we’re just really deep into semantics. I think what she’s trying to get at is some detail about the difference between a set of sensations for which we have no mental label, and a set of sensations for which we have a mental label, but I don’t feel like I quite get it.

And while I think it’s useful when talking about what we thing is happening neurally, I’m not sure this is a useful way to talk outside of this context. When my 3-year old doesn’t get what she wants and starts screaming and flailing her arms is it useful to say “You’re acting as though you’re experiencing energy-budgeting sensations that in myself I categorize as anger, but since you don’t have a solid concept of anger I know it’s not actually anger you’re experiencing”? No. I say “Whoa, you’re very angry!” or perhaps “Seems like you’re angry!”, to which she’ll reply “NO I’M NOT!” – oh! She’s right! This whole time she’s had a better understanding of HEAM than me!

Words are a requirement

LFB seems to believe that words a requirement for developing a concept, or maybe just for developing shared concepts?, and I don’t think that’s true.

The example that pops into my head is noticing a pattern in how someone treats you: It may not be a single type of action, it could be a variety of unrelated things they do and those things might make you feel all different ways, but they feel connected. The only real connection between the things is that they involve a particular person and they feel connected (other things they do don’t feel like part of this pattern). I wish I could remember more details of when I’ve felt this way, and maybe it was just a part of being a highly self-conscious teen, but it was (I think, by LFB’s definition) a concept for which I had no words.

Even for developing shared concepts, I suspect we can do it (just less efficiently) without words. I can’t remember if it was in HEAM or somewhere else, but I read a description of a study where people looked at patterns of dots that were variations on some original pattern. After training on some examples corresponding to different original patterns, they were able to look at new examples and correctly match them to the group of variations they belonged to. This is supposed to imply that people had some mental concept of the original pattern, even though they’d never seen it. So imagine you’re doing this activity silently with a friend. You have no words for the groups, but you’re both able to categorize these patterns in the same way – I’m pretty sure that’s a shared concept?

I’m probably misunderstanding some nuance of what she’s saying.





When combined with ideas from Atomic Habits (TODO: write up a page for that book), TODO…

Last modified: 2023-01-28 18:46:57 -0500 -0500 • For full version history, see github
Tags: book