The layers of emotion creation

Image credit: Toddatkins via Wikipedia

What are emotions?  Where do they come from?  Are they something innate or something we learn?  The classic view is that they’re precognitive impulses that happen to us.  If so, this would imply that they have specific neural signatures.

Early in her career, psychologist Lisa Feldman Barrett attempted to isolate the neural basis of emotions using brain scans, but discovered no consistent patterns for any one emotion.  For example, the neurons involved in anger were not consistent in every brain, or even in every instance of anger in a particular individual.  No individual neuron consistently fired in every case.

This evidence seemed to line up with what she had already found when attempting to find the behavioral signatures of emotions.  After discovering problems with research showing multiple cultures the same emotional faces, she eventually discovered that reading real emotions via facial expressions has little or no objective basis.  It only seems to work when the other person is from a similar culture.  Objectively reading emotions from the faces of people from other cultures proved to be impossible.

Each emotion, she found, was more a category of neural activity rather than a discrete process.  Their diffuse nature led her to conclude that emotions are mental concepts.  Similar to other types of concepts such as chairs, dogs, or hiking, they are things we learn, sensory predictions our brains develop over a lifetime.

Somewhat in opposition to her understanding, is the one championed by the recently deceased neuroscientist Jaak Panksepp.  Focusing on animal studies, Panksepp identified what he referred to as seven primary emotions, labeling them in all caps as: RAGE, FEAR, SEEKING, LUST, CARE, PANIC/GRIEF, and PLAY.  Panksepp identified these primary impulses as sub-cortical and mostly sub-cerebral circuits arising from the brainstem and mid-brain regions, deep regions that are difficult to reach with brain scans.

So who is right?  As is often the case, binary thinking here can mislead us.  I think it helps to review the details on how they agree and disagree.

Barrett admits in her book, How Emotions Are Made, that most animals have circuits for what are commonly called the “Four Fs”: fighting, fleeing, feeding, and mating.  It doesn’t seem like much of a stretch to link Panksepp’s RAGE to fighting, FEAR to fleeing, LUST to mating, and to see feeding as a result of a type of SEEKING.  In other words, Barrett’s admission of the Four Fs gets her halfway to Panksepp’s version of innate impulses.  It’s not hard to imagine that mammals and birds have innate impulses for CARE and PANIC/GRIEF.

However, there is a big difference in how Barrett and Panksepp label these circuits.  Barrett makes a sharp distinction between emotions and affects.  She regards affects as more primal states and characterizes them in a fairly minimalist manner, sticking to the traditional description with two dimensions: valence and arousal.  Valance is an assessment of how good or bad a stimulus is.  And arousal is the level of excitement the stimulus causes.  Barrett either is unaware of or disagrees with the idea put forth by other scientists that affects also include an action motivation.  She sees affects as an early pre-emotional stage of processing.

Panksepp in contrast, in his own book, The Archaeology of the Mind, seemed to view the words “emotion” and “affect” as almost synonymous, or perhaps saw emotions as merely a type of affect.  Panksepp’s view was that the primal drives he identified were feeling states, that they include some form of lower level consciousness in the sub-cerebral regions he discussed.

In this sentiment, Panksepp stood in opposition to Barrett’s view that these are best described as “survival circuits”, non-conscious reflexes.  From what I’ve read, the vast majority of neurobiologists would agree with Barrett on this point, that consciousness is a cortical phenomenon, at least in mammals, and that sub-cortical, or at least sub-cerebral structures are primarily reflexive in nature.

Interestingly, both Barrett and Panksepp push back against the popular notion that emotions, particularly fear, originate from the amygdala.  In Panksepp’s view, the amygdala seems like more of a linkage system, linking cerebral memories to the primal impulses he identifies as coming from lower level structures.  Both acknowledge that patients with destroyed amygdalae can still experience fear, although generally only in response to a subset of the stimuli that cause it in healthy people.

And Panksepp admits early in his book that social learning has a major role to play in the final experience of felt emotions in the neocortex, which seems like a major concession to Barrett’s views.

One significant difference between them, which may come down to a definitional dispute, is whether animals have emotions.  Panksepp, using his understanding of emotions, unequivocally considered animals to have them.  Barrett, while admitting some great apes may come close, doesn’t think animals have much beyond what she considers affects.   In her view, most species lack the neural machinery to learn emotions the way a human does.

My own reaction to all of this is similar to how I view consciousness, as a multi-level phenomenon.  It seems pointless to argue about how and where emotions arise unless we can agree on definitions.  I perceive most, though not all, of the disagreement between Barrett and Panksepp to amount to a difference in definition, in what is meant by the word “emotion”.

Regardless of how we label them, it seems like what we experience as emotions start as lower level reflexes in sub-cerebral regions, which reach cortical regions as primal feelings.  Over a lifetime, the interpretation of these primal feelings change and evolve into what we commonly refer to as emotional feelings.

It seems like there’s a major divide here between interoception, sensory perceptions from the body, and exteroception, perceptions from distance senses such as sight, hearing, and smell.  We probably have strong innate primal responses to certain interoceptive sensations, but mostly have to learn which exteroceptive ones to link to those primal responses.

In other words, we don’t have to learn that a stinging sensation on our hand is bad.  We likely know that innately.  But we do have to learn (as I once did as a small child) that sticking our hand in an anthill will result in the bad stinging sensation.  Once we’ve done it the first time, our amygdala will link the memory of an anthill to the negative valence of a stinging hand.

So emotions, it seems to me, have an innate core and a learned response.  Crucially, introspectively separating the learned response from the innate primal feeling is extremely difficult, perhaps impossible.  This is why anthropologists can often say that emotions are not universal across all cultures.  In their final learned form, we shouldn’t expect them to be.

What do you think?  Do you see any issues with either Barrett’s or Panksepp’s view?  Or my own syncretized one?  Am I missing anything?

43 thoughts on “The layers of emotion creation

  1. This is all very much a work in progress and possibly too preliminary to judge. It is very interesting though. babies are born, apparently, with a fear of falling and loud noises, both of which create a startle response and usually loud crying. hardly the response you would want if either were the result of a predator’s actions. I imagine that evolutionary science and neurobiology and psychology are all going to converge sometime in the near future on this topic. We have apparently unlocked the mechanism behind facial recognition, so these even knottier problems will succumb, and soon, I think.

    Liked by 1 person

    1. I agree it’s a work in progress, but that seems like a statement we can make about any scientific enterprise. But unless Barrett’s or Panksepp’s work is flawed (admittedly always a possibility), any new understanding will have to explain their results.


  2. I was under the impression emotions were just little people in our heads fighting over a control panel. At least that’s what I learned from the movie “Inside Out.”

    It is interesting to me (and I think you can guess why) that so much of this depends on how we define our terms. Admittedly I’m not an expert on psychology, but it seems to me that emotions are very subjective things, and I’m not sure if we can ever fully describe them in an objective way.

    Of course, once again, I’m not an expert on this at all. So perhaps I’m the one missing something.

    Liked by 2 people

    1. 🙂 I haven’t seen the movie ‘Inside Out’, but I’ve heard it mentioned several times by psychologists and neuroscientists, some approvingly. Barrett mentions it but as an example of propagating what she calls the “classic” model of emotion, as something that happens to us.

      Based on Barrett’s findings, I think you’re right about the strong subjective aspect. Using her distinction, affects may be more objective (although it sounds like even they are subject to alteration by learning), but how we interpret them into emotions is probably unique for every person in their lived experiences and culture. We classify certain categories of reactions as being “anger” or “happiness”, but it’s an interpretation of the underlying reality.

      Liked by 1 person

  3. Never heard of Barrett till today, and then it happened twice. I first saw her on BookTV earlier (at a bookstore I frequent!) and now your excellent post.

    There were two salient points that I took away from her informative and entertaining presentation. First. She subscribes to a computational model of consciousness and nicely described how everything in our awareness is an internal process in which all the events in our body are monitored by the nervous system and then presented to our awareness as representations/simulations. Second. Simulations of emotions are not that different from other simulations such as color and smell. She did make a big point out of the benefits to be derived from having an improved emotional vocabulary – learning and communication become so much more efficient when we can specifically name an emotion . ( I would say that the limitations of our vocabulary, especially as it relates to our subjective experiences, is a major obstacle that we have in understanding and communication on these matters, and others obviously.)

    Liked by 1 person

    1. Thanks Liam!

      That sounds like it was an interesting Book TV talk. I might have to see if I can find it. Her book’s been out for a few months now, but it seems like there’s been a recent surge in publicity for it. She also recently did an interview on the Brain Science podcast with Ginger Campbell:

      In case you’re interested, Campbell also recently re-released an interview she did with Jaak Panksepp after his passing was announced:

      Totally agree that finding the right vocabulary, particularly in terms of emotion, is often challenging, and limiting when it isn’t found. And definitely any discussions about the mind, consciousness, and the brain are constantly hampered by the limitations of language.

      Liked by 1 person

  4. The talks about the machinery required for the brain to process emotions, make me a little skeptical. What if there is something too tiny to discover and too usual to be ignored? There should be specific areas in the brain for specific tasks like language processing and storage of certain memory, which are there. But the major four emotions of, fight-flight-feeding and mating, must be a part of every single unit of brain, i.e every single neuron.

    If emotion was a resident of one or few specified part of brain, it would be very difficult for the being to sustain. But if emotion was everywhere in the brain (distributed unequally though), the fact that even after removing the amygdala the patient feels fear, is very much possible.

    Liked by 2 people

    1. Thanks for your comment Vaidikakashikar, and welcome!

      On the Four Fs, I think the answer is that they exist in circuits at the base of the brain, the evolutionarily oldest parts, where they can influence the processing of many other regions. The brain has multiple redundant pathways for many of its functional networks, which is how fear can still happen without an amygdala, although this isn’t absolute since people with destroyed amygdalae fear only a small subset of the things that healthy people do.

      But to your broader point about emotions in animals, I think this is why I prefer conceiving of emotion as a multi-level process. Doing so allows us to acknowledge that a dog feels a certain level of emotion, while leaving room to also acknowledge that it’s not at the human level. So a growling dog is experiencing agitation, but it’s probably not accurate to say it’s angry in any way like the human version of anger.

      Liked by 2 people

      1. Thank you sir, Mike (as I get to know that here) and James!
        May be we can’t really locate the exact organelles for producing and regulating emotions, we can anyway make sure that the chemical factors affecting emotions are working well. Isn’t that another paradox that when drugs are consumed in order to increase the rate of serotonin to get a happy feeling, with time it lessens the number of receptors throwing the brain back to the same need again. This is ironical!

        Liked by 2 people

    2. I agree, mostly, with Vaidikakashikar. I always assumed (so no evidence here) that emotions were nature’s way of generating global influences over the body, either by hormones (like adrenaline) or by neural extensions that are excitatory or prohibitive, but only slightly influential on their targets.

      Liked by 3 people

  5. Maybe it’s not so helpful to treat emotions as if they were of a different category to thoughts and feelings? I see an emotion as a specifically natured thought arising conascently with an altered (from the neutral norm) mental feeling. It isn’t anything else. All thoughts give rise to mental feelings (mood tones), even if those feelings are neutral i.e. neither pleasant nor unpleasant. The feelings become variant to the neutral norm dependent either upon an arisen thought or some physical sense contact. In short, for me an emotion is two events conflated (actually confused) as if one. Because of this view, I wonder if it’s not rather redundant looking for places in the brain where emotions arise — what’s looked for is something that likely has no unified source or locale, because it’s two processes (mood tone and ideation). Good article, Mike, as always, and for which many thanks.

    Liked by 2 people

    1. Thanks Hariod. Your terminology is different (being more eastern and spiritual perhaps) but it seems like you’re making the same distinction Barrett does between affects and emotions, with affects being the mental feelings and emotions the natured thought or ideation, assuming I’m understanding correctly.

      It can be productive to look at this in various ways. I can actually see three distinctions: raw interoceptive sensations, affective feelings, and emotional experience. The first depends solely on the state of the body, the second is heavily effected by the body state but adds valence, arousal, and instinctive motivation, and the last relates it within a galaxy of associations that we learn over our lifetime. The final result seems like a complex synthesis of all of them, one that seems subjectively irreducible.

      Liked by 1 person

      1. On the contrary, I don’t see anything remotely ‘spiritual’ in what I say, Mike. An emotion is a label we put on a subjective experience which itself consists in nothing more than thoughts and feelings arising conascently as an apparently unified, meta-level representation. Naturally, those thoughts and mental feelings (mood tones) are themselves conditioned by a myriad other things — the weather, what side we get out of the bed, whether our team lost yesterday, etc. The question is whether it’s valid or productive to selectively include just some of those other things (your ‘galaxy of associations’), none of which are extant within the purely subjectively experienced emotion itself; they’re causally associated, but not themselves present in the arisen emotion. Yes, an emotion is ‘subjectively irreducible’ (other than as thoughts and feelings), because the emotion (by my and most common definitions), is defined purely by the experienceable, not the myriad causal associations of it — Toyota’s robots are not part of the Avensis they built for me. If we start saying the robots are part of my car, then we end up going back to the beginning of time, constantly moving away from what an Avensis actually is, in and as itself, and the concept of ‘Avensis’ becomes meaningless.

        Liked by 3 people

        1. Hariod, the only reason I said “spiritual” was that when I googled some of the words you used, such as “conascently”, a raft of Buddhist and mindfulness material came up. Sorry if I misclassified it. It absolutely wasn’t meant in any judgmental way.

          I’m not sure if I see the Toyota analogy. It seems like a comparable notion would be if I got into the embyronic and fetal development of the brain as a cause of the emotion. Certainly there are causal factors from all directions, but it still makes sense to understand them to the extent we can.

          A real world example that Barrett used in her book were parole judges. Studies show that they turn down more candidates for parole just before lunch, likely mistaking their affective hunger signals for gut level intuitions about the trustworthiness of the candidates. The approval rates go back up immediately after lunch. Yes, there are innumerable other factors involved, but understanding how their interoceptive and affective states influence their emotional and rational ones seems like a crucial insight.

          Liked by 2 people

          1. How strange that a standard variant of ‘nascent’ should show up in Google’s results with Buddhist links — I’ve no idea why; it’s in all the standard dictionaries without the ‘co-‘ prefix. You know what I mean though, Mike: coming into existence together in the incipient stages. So as I see it, the emotion only exists at this point onwards, which is as a thought and associated mental feeling begin to arise within and as consciousness, not at any prior causal point.

            With that in mind, and clumsy though the analogy admittedly is, to talk of what an emotion is prior to this point is like asking what my Avensis was before it existed — and that isn’t just limited to what happened in Toyota’s production plant. I guess we’re back to that fundamental difference we have on consciousness; simply put: you tend to see it as something occurring in the brain, and myself as something broader based involving what is external to the body.

            Your first question is, What are emotions?, and your second, Where do they come from? I think I’m providing a reasonable enough answer to the first, albeit a simple one, in that an emotion is a conascently arisen thought and mental tone. The second question seems impossible to answer without putting limits on where the search ends, which itself nullifies the question. If we say an emotion’s causes begin and end within the body/mind then I think that’s incorrect, as would you, I think?

            Liked by 2 people

    2. Hariod,
      I did eventually figure out “conascently”, but it took a bit for my befuddled brain to figure out it was co-nascent. Once there, I think I got your meaning. 🙂

      On where the emotion exists, my inclination is to use the word to refer to the combination of the affect and the concept. Often it also includes the interoceptive aspect, although while it has a enhancing or inhibiting effect, I’m not sure it’s essential. But if you remove the affect, it’s hard to see how the remaining concept is what we think of as an emotion. And I’m not sure if we can coherently talk about removing the concept, since we only seem to know the affect through the concept, although we also feel its interoceptive effects.

      (Of course, James-Lange theory posits that we construct emotions purely on what we feel interoceptively, but I think most neurobiologists think there is direct information transfer from the affect circuitry to the concept circuitry. The connections certainly seem to exist for it.)

      Granted, it doesn’t make sense to include the stimulus, interoceptive or otherwise, or the imaginative scenario, or whatever caused the affect and concept, to be part of the emotion. I guess the reason I resist putting the affect in the cause category is because it can go both ways. If we trigger the concept, the affect seems to follow. Think about something that makes you angry, and after an anger concept forms, it will usually lead to the associated affect, albeit at a lesser intensity than if you were currently experiencing the situation.

    3. I think this is similar to what Lisa Feldman Barrett is saying. That basically we have these physiological responses to a spectrum between pleasant and unpleasant, and between arousal and calm. That’s it. And that these responses then get interpreted as emotions, But it’s much more like saying…well when does red become red? And am I interpreting red the same way as another person might interpret red? And maybe I’m misinterpreting what you’re saying, but I think that Barrett would say yes we experience the tone, and the ideation is learned. So what we know what to call it. I mean we experience seeing red, before we know to call it red. But red is just a name, and it might apply to various shades of red, and what defines red to us is what we are taught by our parents and others as red. In this way as children have the mood tones, they don’t know what to call those feelings, but I would imagine that this is learned from others as well as the language we’ve developed to describe those feelings which is the ideation.

      I think this is also why children can confuse emotional states with certain actions. If you had parents who were unkind to each other and fighting a lot or unkind to you, what you identify as feelings of love, may lead to expressions of love that are more dysfunctional, using your parents as a working model on how to act on the mood tones you’re experience.

      Liked by 1 person

  6. Hi MIke,

    Great post, once again, which I enjoyed reading. I always learn something here. I had a moment of resistance to the idea that an emotion could be placed on par with a “concept” such as a chair. From a subjective personal perspective, I see emotions as being quite different. I’ve thought about it and decided that emotions seem to be thoughts–(not sure what word to use besides thoughts, though the emotion often doesn’t feel as though it coincides with any single thought, but rather seems to fuel a cascade of related thoughts whose origin can eventually be found by carefully sifting through what is thrown up on the mental screen)–anyway, emotions seem to be thoughts I experience through my body, whereas using my imagination to picture a chair can be done without any other bodily sensations whatsoever. What I think of as emotion always (I think) involves bodily feedback. It’s hard to be angry without feeling it in my body. Likewise I experience sadness in the body. We don’t cry intellectually, right? I suspect this is true of other emotions as well.

    I’m curious what you’ve gleaned from these sources, if anything, about this idea? Do you think an emotion is the same as a concept?


    Liked by 3 people

    1. Thanks Michael! Glad you enjoyed it.

      Barrett definitely equates emotions with concepts. At a certain level and in a certain manner, I agree with her. But as you noted with the chair, not all concepts are emotions, or have emotional resonance. I think emotions are concepts that are associated with certain affective and interoceptive states. Barrett downplays this relationship, but I think it’s definitely there. In other words, I agree with you on the bodily feedback aspect.

      Of course, the relationship is complicated. An interoceptive stimulus can create an affective state, which in turn can create an emotional one. But if I imagine a scenario that makes me angry, the activation can run in the other direction. And a body state can change what might have been a minor emotional event into a major one, such as getting angry at a minor snub when being jacked up on too much caffeine, or when in pain.

      Every time I run through this in my mind though, I can’t escape thinking that an emotion is not just the concept, but also the activations up and down the stack. I’m not sure Barrett would agree with that assessment.

      Liked by 2 people

  7. Enjoyed these notes and the commentaries above. I’ve decently skimmed her book but hope to read all of it.

    I think much of her take is probably right. This gets back into the idea that our describing our phenomenological world, our understanding of our minds, arose before a more formal science arose. Even when psychology ramped up it was still in a troubled area about interrogating the mental world, but also those mental worlds had inculcated poorly thought theories back into us. We started seeing our selves within these poorly described ideas and this had additional effects on us. That is, once these these words and ideas were created they became ways to think about our own selves and to guide behaviors.

    As we start looking under the hood, as we start looking at brain scans and engage with a more thorough science, we are realizing the messiness of these brain systems. We are also uncovering how previous conceptions and words about those mental states were glossing over significantly differing brain/mental phenomena.

    Anyways, that’s my reading. I’ve read several of Damasio’s books and still struggle with emotional concepts, though I do like his emphasis on bodily monitoring. I would like a good accounting of pain (and pleasure) and I think we can put the whole mind thing to bed.

    162 “Essentialism also appears to be an inherent part of our psychological makeup. Humans create categories by inventing purely mental similarities . . . and we name those categories with words. That’s why a word like ‘pet’ or ‘sadness’ applies to a multitude of diverse instances. Words are incredible achievements, but they are also a Faustian bargain for the human brain . . . That is, the word sadness guides you to create an emotional concept, which is a good thing. But the word also invites you to believe in a reason for that sameness.”

    One last thing, and you may have posted this before, but there was an interesting paper recently by Richard Brown and Joseph Ledoux on emotional consciousness:

    Skimming back through that, they seem to agree that human fear is both a sub-cortical more basic response as well as a channeled social response. They go through higher thought theory as to how emotions become known consciously.

    Liked by 1 person

    1. Thanks Lyndon!

      The idea that our theories have modified us is interesting. I can definitely see it setting up a confirmation bias in what we notice about our mental processing or behavior. But this also gets into the enormous influence that culture has on how we think, particularly in how we conceive ourselves and our minds.

      I’ve only read one of Damasio’s books (‘Self Comes to Mind’), but I sometimes wonder if I should read ‘Descartes’ Error’, just because it gets cited so often. But I definitely have to credit Damasio with shaking me out of what were, in retrospect, fairly simplistic notions of how consciousness and minds work.

      Thanks for linking to the paper! I didn’t think I’d seen it before, but as I started reading it, I realized I’d seen the HOR, HOT, and HOROR concepts before, possibly in this paper. Scanning the abstract, their thesis seems very similar to Barrett’s, that emotions are high order cortical phenomena, with subcortical circuits essentially being non-conscious inputs.

      Their distinction between first order theories of consciousness and HOT (higher order theories) is very interesting, particularly in pointing out that Panksepp, and to a lesser extent Damasio, are essentially drawing on a first order understanding of consciousness, that is a version of consciousness that exists independently of introspection, even the passive variety.

      I’m slowly coming to agree with them that first order consciousness is probably a mistaken concept. Subjective experience requires at least passive introspection. Without it, whatever is going on is non-conscious. I find this conclusion unpleasant due to its implications for consciousness in most animal species, but it may be reality.

      I haven’t read the full paper yet but plan to keep working my way through it.


  8. I found it hard to believe that “she eventually discovered that reading real emotions via facial expressions has little or no objective basis. It only seems to work when the other person is from a similar culture.” So I did some Googling, and now I find it even harder to believe. Some top results for “intercultural recognition of emotion” are Soto and Levenson and Sauter et al. The former say “Our findings for empathic accuracy supported the cultural equivalence model, with no evidence of greater accuracy when [American] raters viewed targets of their own ethnicity.” The latter say “Western participants were compared to individuals from remote, culturally isolated Namibian villages. [Nonverbal] Vocalizations communicating the so-called “basic emotions” (anger, disgust, fear, joy, sadness, and surprise) were bidirectionally recognized. In contrast, a set of additional emotions was only recognized within, but not across, cultural boundaries.”

    Liked by 2 people

    1. Barrett discussed the classic experiment from the 1960s by Silvan S. Tomkins where photos of the facial expressions of various actors showing various emotional states were shown to subjects in several cultures. People were apparently able to accurately assess the emotional state in the pictures. However, these were actors and the subjects were interviewed by western interviewers using western concepts.

      She continued with this:

      Other scientists, however, worried that the basic emotion method was too indirect and subjective to reveal emotion fingerprints because it involves human judgment. A more objective technique, called facial electromyography (EMG), removes human perceivers altogether. Facial EMG places electrodes on the surface of the skin to detect the electrical signals that make facial muscles move. It precisely identifies the parts of the face as they move, how much, and how often. In a typical study, test subjects wear electrodes over their eyebrows, forehead, cheeks, and jaw as they view films or photos, or as they remember or imagine situations, to evoke a variety of emotions. Scientists record the electrical changes in muscle activity and calculate the degree of movement in each muscle during each emotion. If people move the same facial muscles in the same pattern each time they experience a given emotion— scowling in anger, smiling in happiness, pouting in sadness, and so on— and only when they experience that emotion, then the movements might be a fingerprint. 9

      As it turns out, facial EMG presents a serious challenge to the classical view of emotion. In study after study, the muscle movements do not reliably indicate when someone is angry, sad, or fearful; they don’t form predictable fingerprints for each emotion. At best, facial EMG reveals that these movements distinguish pleasant versus unpleasant feeling. Even more damning, the facial movements recorded in these studies do not reliably match the posed photos created for the basic emotion method.

      Barrett, Lisa Feldman. How Emotions Are Made: The Secret Life of the Brain (pp. 7-8). Houghton Mifflin Harcourt. Kindle Edition.

      Liked by 2 people

      1. Thanks. It seems we can say two things. One, actors asked to clearly (i.e. extremely) display certain emotions can succeed in conveying the nearest translation-equivalent to people worldwide. Two, the muscle movements are not so blatantly obvious that even a nearly-modern computer program can infer emotions from crude muscle measurements.

        But as someone who regularly works with getting computers to recognize categories based on limited information, I can assure you that they regularly underperform humans (who typically access more info) at the classification task. And if there’s one task at which humans should be able to kick computer butt, recognizing human emotions is probably it.

        Liked by 1 person

        1. Good points. I actually had an ending paragraph in the original comment above discussing some of the caveats, but WordPress appears to have eaten it for some reason. That paragraph discussed Barrett’s caution that the EEGs conceivably might not have been granular enough (people get uncomfortable with too many electrodes on their face). She discussed an alternate method called FACS (facial action coding), which found similar results, but allowed some human judgment to creep back in for the coders.

          It’s the corroboration with the failure to find consistent neural fingeprints for each emotion that give these results some credibility. If the neural firing patterns aren’t consistent, it’s not hard to imagine that the facial muscles or other behavior signs wouldn’t be as well.


  9. Hi again Mike and others,

    Just finished watching the HBO series Westworld tonight and was curious if you had seen it, since you post on Scifi. I did not see it in your search.

    Their exploration of consciousness was well done, especially using the bicameral mind notion (Julian Jaynes) and the nature of memories. Their meandering with emotion and feeling might be more strained. My take is that such creatures are clearly conscious, with robust self and world models and what seems like significant pain and emotional responses. But the world of programmed and carefully curtailed AI will be a strange place. Though, the AIs robust improvisation also seems like it would guarantee consciousness combined with their other attributes.

    To achieve AI consciousness: Memory; Improvisation; and then create and bootstrap the bicameral mind.

    Anyways, surely the best discussions on consciousness within popular TV, and enjoyable to boot.

    Liked by 1 person

    1. Hi Lyndon,
      I did watch Westworld and enjoyed it very much. I didn’t find its musing on consciousness, in and of themselves, particularly insightful, and although I’ve never read Jaynes’ book, I can’t say I find much stock in the bicameral mind concept. But in addition to being entertaining, Westworld was fascinating for what it seemed to reveal about our intuitions about consciousness, how important human instincts and emotions are to triggering our intuition of another consciousness.

      I agree with you there’s no way the hosts could have done what they did, behaved the way they did, including all the social interactions and emotional cuing, without being conscious, at least when they weren’t in diagnostic mode.

      I think the most interesting moment in the series for me was this exchange (via IMDb –

      Bernard Lowe: So what’s the difference between my pain and yours?

      Dr. Robert Ford: Between you and me? This was the very question that consumed Arnold, filled him with guilt, eventually drove him mad. The answer always seemed obvious to me. There is no threshold that makes us greater than the sum of our parts, no inflection point at which we become fully alive. We can’t define consciousness because consciousness does not exist. Humans fancy that there’s something special about the way we perceive the world, and yet we live in loops as tight and as closed as the hosts do, seldom questioning our choices, content, for the most part, to be told what to do next. No, my friend, you’re not missing anything at all.


      1. That was a good quote. I was going to paraphrase that yesterday.

        I agree on the bicameral mind, I was just impressed with its use as a plot device.

        Their denial of consciousness to AI does show one thing, we have to have the right kind of self representation to be aware of our self, to be fully conscious. First robust representations of self and world, but then maybe some kind of recursive knowledge, “I know that I am aware of my self.” I would say that last step is something that arrives through language and other modeling, and is what keeps animals and babies in the dark, so to speak. I guess the plot of the show is that you could have very impressive AI and they still may not model or represent that last piece of consciousness, but as you said, that is probably dubious.

        Liked by 1 person

  10. Hi Mike, great article. I think in general I side with Barrett’s view of separating affect and emotions. Then again as you say it may come down to the way we define each and that seems to have always been a subject of confusion and debate (like the Lazarus-Zajonc debate in the 80s).

    On definitions, whilst reading this post I had some (?epistemic) ideas/question;
    Emotions are a subjective experience and there seems to be a problem in their objective definition. [A]: If the involved brain activity is different in 2 individuals then why do we say they both felt the ‘same’ emotion for example ‘anger’? (i.e. how do we know it’s the same thing).
    [B]: Perhaps the way we do it is by the similarity in its objective manifestation. Yet even the objective expression is never the same in 2 individuals. This difference seems to be more pronounced in individuals from different cultures. So if not the objective expression how else do we categorize both as being the same emotion?.
    [C]: It is perhaps because the individual can tell us that what he feels is ‘anger’. But how does the individual know what to call what he is subjectively feeling? He has to have learned what this feeling is. Yet this learning will be flawed if it is based on the above i.e. [A] and [B].
    [D]: Taking 2 instances of anger within the same individual. Again if the objective correlates of it in terms of brain activity and behavioral manifestation are different each time, why should it be the same emotion? We can’t trust what the person says he feels based on [C].
    [E]: It also seems the person isn’t in the same state in both instances. He ‘thinks’ both were the same but this isn’t the same as feeling them the same. Thinking involves thought, this will again be flawed based on his flawed knowledge as in [C].

    May be that’s part of the problem in defining emotions.

    Liked by 1 person

    1. Thanks Fizan!

      Your lettered points are pretty much the same ones that Barrett makes. Her conclusion is that an emotion, such as anger or fear, isn’t a discrete thing, but a category of mental activity. Like all categories, there are similarities between the individual instances. They overlap with each other enough for us to recognize them as similar. In brain scans, similar regions may light up, but no one neuron or group of neurons will light up in every case of that emotion.

      She compares emotions to animal species. There are wide variations between members of specific species, and sometimes substantial judgment is needed to determine whether two individual animals are members of the same species. This is also similar to how we group sections of the color spectrum into specific named colors, with different cultures dividing the spectrum in different ways. Everyone sees a similar spectrum, but the labels vary.

      My issue with separating the emotional concept from the underlying affect, is that if you remove the affect, I’m not sure we’d call what remained an emotion. If something triggers the emotion concept, it usually triggers the underlying affect(s).

      But admittedly it’s complicated. There isn’t a one to one relationship between affects and emotion concepts. The same affect can trigger multiple emotions. Often an emotion concept is essentially an interpretation, what Barrett calls a prediction, marrying an affect to a particular context.

      Liked by 1 person

      1. On reflection, although I do side with Barrett in making a distinction between emotions and affect it perhaps isn’t the same type of distinction she makes (not to say I disagree with her making a distinction). The word ‘Affect’ appears to have been misused a lot with different experts using it in very different ways, whilst emotion seems to be a more stable concept. In psychiatry I studied three ways it has been used:

        As ‘Affective’ Disorder – These are also confusingly called ‘Mood’ Disorders. The DSM 4 defined Affect as the visible and audible manifestations of the patient’s emotional response to external and internal events.
        Emotion theorists (like Barrett) and some phenomenologists – Use it as the non-observable ‘feeling part’ of emotions or the moment to moment ‘feeling state’.
        Common usage – As all things emotional i.e. it is an umbrella term which encompasses all related things like sentiment, preferences, emotion, mood and affective traits etc.

        In all, three usages it is considered different from emotions. The last usage is the one commonly used (that is not to say the other differences aren’t significant).

        Clinically we also distinguish between affect and mood. Mood is the longitudinal/longer-term state of Affect, which is the crossectional/current state.

        Liked by 1 person

        1. I agree with you about the word “affect” being used in many different ways. Just in the neuroscience literature, affects are sometimes the non-conscious survival circuits, or they are primal emotional feelings, or they are the displays you describe. And writing about affects is difficult because people often confuse the noun with the verb.

          But I’ve also found similar issues with “emotion.” Maybe it’s more consistent in the psychological literature, but in the neuroscience and philosophical literature the word can mean anything from an instinct (similar to the survival circuits for affect) to the complex social psychology concepts Barrett describes.

          All of which is why I usually write about reflexes, which most people understand are non-conscious lower level circuits, and feelings, which are generally conscious states. Although even this can be complicated by the fact in that it seems like we can have unconscious feelings which effect our behavior.

          And I’m very aware that the word “reflex” has its own problems. A mid-brain reflex isn’t like a lower spinal cord reflex, in that it can be inhibited, and has been co-evolving with the higher level systems for a long time. This has made them less specifically action oriented and more generally priming in nature than the lower spinal reflexes. The word “affect” might be a good word for these circuits, but it has the problems discussed above.

          All of which is to say, language is a pain!

          Liked by 1 person

  11. The idea that affect and emotion are different is not Barrett’s. It’s 150 years old, dating back to Wilhelm Wundt, who helped found psychology as a science. Affect is raw feeling, like pleasant, unpleasant, calm, aroused. See also James Russell’s affective circumplex.

    Liked by 1 person

    1. Thanks for your comment, and welcome.

      I don’t recall Barrett making any claim to being the first person to conceive of this understanding. Her chief approach is describing the scientific data. At this point, it seems like we can find prior writing in support of (or that we can construe to be in support of) just about any theory about the mind.


  12. Hi Mike, I’m back from a holiday and catching up on my reading…

    When I was going through my steps of how to “know thyself”, the theory for emotions that made the most sense to me was “appraisal theory.”

    I pulled together this short summary after reading about it:

    “An influential theory of emotion is that of Lazarus: emotion is a disturbance that occurs in the following order: 1) cognitive appraisal – the individual assesses the event cognitively, which cues the emotion; 2) physiological changes – the cognitive reaction starts biological changes such as increased heart rate or pituitary adrenal response; 3) action – the individual feels the emotion and chooses how to react. Lazarus stressed that the quality and intensity of emotions are controlled through cognitive processes.”

    Think about it; you can’t be scared of that spider behind you until some part of you senses it and cognitively appraises the danger. The intensity of that feeling grows or diminishes usually in direct proportion to the size of the spider too. : ) So there must be a body-mind feedback loop. (Recognizing the mind is really a function of the body too…) I took this idea and ran with it logically, seeing that you can appraise a situation either positively, negatively, or as uncertain. You can also appraise the past, present, or future. When I tried to put a list I found of 70 of the most common emotions into categories using these variables, I found a few discrete situations in the past, present, or future that made sense to lump several emotions together along a spectrum of variable intensity. For example, if I cognitively appraise a “past” event “negatively” where “I acted poorly”, I will feel: regret -> guilt -> embarrassment -> remorse -> shame -> humiliation depending on just how negatively I appraise that event. We also have emotional responses about what to do about fixing negative appraisals. For example, if I cognitively appraise that “I need to do something about that”, then I will feel: determination -> annoyance -> frustration -> anger -> hatred -> mania -> rage in some sort of order. This all gave me a table of emotions and their causal explanations that has proven very reliable and helpful for me over the last few years as I’ve tried to observe emotions in myself and others, and then change the cognitive appraisals where possible if I need to manage my emotions better. You can see my full table and a brief post about it here:

    Maybe you’ll find this useful too. I’d love to know what you think. Thanks.

    Liked by 1 person

    1. Hi Ed,
      Good hearing from you!

      Thanks for the links. Appraisal theory strikes me as obviously true. And it seems broadly compatible with Barrett’s views on emotion. I think the only part that would make me nervous is if it was presented as a one way process. It seems like activation can come two ways:

      exteroception->appraisal->affect->physiological effects->interoceptive feedback


      physiology->interoception->affect->appraisal->perhaps creating a memory linking to current exteroceptions which would later enable the first sequence.

      In many ways, what we call the emotion could be considered the nexus of exteroception, affect, interoception, and memory. Barrett considers an emotion to be essentially a mental concept, a prediction in its own right. The common denominator might be that the appraisal is the emotion, although without the affect, I’m not sure if we would really consider it to be an emotion.

      Liked by 1 person

      1. Hmm. I didn’t know what “affect” meant so I searched through your comments and saw where you explained the difference between Barrett and Panskepp on this. I’d side with Panskepp, and say the “pre-emotional” affect Barrett is describing is just an unconscious emotional reaction. (And aren’t they all, unless we become mindful of them?) I dunno. I’d have to actually read Barrett to understand and say more on this. Then, I was trying to decide how interoception would be separate from exteroception in the causal chain of emotions. I think they might both just be physical changes that the body appraises (consciously / unconsciously; system 1 / system 2). When you talk about memory linking external and internal events, that might be Kahneman’s discussion of system 2’s slow thinking recoding system 1’s fast thinking?

        In my post, I also pulled a definition of emotions as follows:

        “Emotion is a complex psychophysiological experience where an individual’s state of mind interacts with biochemical (internal) and environmental (external) influences. Emotions can be seen as mammalian elaborations of general vertebrate arousal patterns, in which neurochemicals (for example, dopamine, noradrenaline, and serotonin) step-up or step-down the brain’s activity level, as visible in body movements, gestures, and postures.”

        So when you say “Barrett considers an emotion to be essentially a mental concept”, I’m not sure if Barrett tried to divide the mind from the body like a dualist would, or if she’s just focusing on the neurochemical step-ups/downs as “the emotion”. I could buy that as the location of “the emotion” as long as the rest of the processes around those mental changes are understood to be necessary too.

        Liked by 1 person

        1. Barrett is definitely not a dualist. I do think she downplays the role of affects in emotions too much. Similar to you, I’m closer to Panksepp’s views on the primal subcortical circuits, although I think he’s wrong when he characterizes those circuits as including some form of conscious feeling. But I do think Barrett is right that the higher level integration has a major effect on our conscious experience of the emotion.

          On appraisal, I think maybe there’s a distinction worth checking here. Does “appraisal” only mean conscious assessment? Or could it also mean an unconscious habitual assessment? Or maybe even an instinctive reflexive reaction? If all three, then appraisal is definitely true, but perhaps trivially so, since even the most primal affect requires some sort of processing other than just receiving signals. Pain isn’t pain without the affective systems.

          On the variation between exteroception and interoception, I should have clarified that the real distinction is between innate reactions and learned ones. It seems to me that we come pre-wired with more interoceptive associations than exteroceptive ones (which is why I used them the way I did above), although we do have some exteroceptive ones, such as the fear reaction all primates seem to have toward a slithering snake like shape.

          On your definition, the scope of which species experience emotions seems to be an ongoing debate, and hinges on how broad or narrow a definition of emotion we want to use. All animals seem to have circuits for the “Four Fs” (fight, flight, feeding, mating). It’s hard to see how activation of these circuits doesn’t have at lest incipient feeling components in any creature capable of non-reflexive action. But a mouse’s version of the feeling of their fight circuit activation is substantially less rich than a primate’s, or a human’s.


  13. This is the continuation of a discussion between Mike and I from the last post, and specifically in response to the following comment:

    Thanks for the link to the dog/guilt study. The site came up with Cyrillic script, which was a bit sketchy, but did provide a six page PDF regarding the position of Alexandra Horowitz and how her study was done. What struck me right off however is that she didn’t even pretend that she was going to objectively attempt to understand the issues here. Again and again she referred to what she was suppose to be testing for, as “anthropomorphisms”. If anything I’d consider that a potential conclusion to reach, not something to bake into the abstract. Hopefully with the replication crisis scientists will begin displaying a more professional attitude than this 2009 study did.

    I interpret her findings quite a bit differently than she did. Apparently she didn’t consider the potential for a dog that is wrongly implicated by a beloved master, to feel even worse than if it was actually guilty of something. (Did she really believe that ill treated animals would display no emotions about their treatment, and so exonerate the position that she was clearly attempting to disprove?) Anyway these negative emotions, which are different from guilt, might very well look to us like guilt. I say this from personal experience. Many years ago I recall my wife accusing me of doing something horrible, and that she could see from the look on my face that it was true. Since I’d done nothing of the sort I found the whole thing quite shocking. I then had to explain to her that the look on my face was actually “hurt” given that she was hurting, and given that she suspected that I had done such a horrible thing.

    I’m obviously not going to grant Horowitz her conclusion. If science is to overturn the beliefs of the amateurs and professionals who intimately work with animals everyday, far better studies will be needed than this one. It’s interesting that Barrett sited a paper which suggests dog jealousy, and then disqualified it for being sloppy, just as I disqualified this one. In the end I expect us to achieve better studies as well as better models from which to interpret those studies. Therefore you and I will need to be prepared to take such progress however it turns out. It may be that many animals have rich emotional lives, or instead it may be that the human evolved to somehow be “special”.

    I’ve mentioned to you before that I do not accept the “agree to disagree” meme between us. Wherever I’m wrong, it is my desire for you to demonstrate my wrongness. I need others to help me alter or abandon my models and beliefs which conflict with valid evidence, as a service to me. My goal is not to be “right”, but rather “real”. I realize that you’re no different. I’ve only just begun considering this metacognition business, and it does seem like a worthy topic to get my head around. Thanks for providing the wrong paper!

    In the end I suppose that we’ll settle this as a definitional issue. I’m going literal on metacognition, which is to say thinking about nothing other that the concept of “thought” itself. Here metacognition is potentially academic, but not really adaptive in an evolutionary sense. Most seem to use a much more broad definition, or reflection in general. Well yes from that definition, of course metacognition can be evolutionarily adaptive. But in that case I must also say that dogs seem to reflect, and even dream in their sleep. Take away our language and our culture, and I don’t currently believe us to be all that special.

    Liked by 1 person

Your thoughts?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.