A (0:04)

Welcome to Brain Science, the podcast that explores how neuroscience is trying to unravel the mystery of how our brains make us human. This is episode 151 and I'm your host Dr. Ginger Campbell. Today we will be talking about the neuroscience of emotion, but before we get started I want to remind you that you will find complete show notes and episode transcripts and you can send me email@brainsciencepodcastmail.com after the episode. I hope you will check out the show notes because they will contain links to many of our previous episodes about emotion. Also, don't forget to send me an email if you're interested in going with me on my trip to Australia in May 2019. And please listen all the way through to the end of this episode because I want to tell you about the new podcast I just launched a few weeks ago. This episode is largely based on the book the Neuroscience of a New Synthesis by Ralph Adolphs and David J. Anderson. I'm always on the lookout for good, good books about emotion because I think it's one of the most fascinating areas of study. This book is rather unusual in that Adolfs and Anderson do not propose any sort of theory of emotion. Instead, their goal is to give an overview of the neurobiology of emotion and to propose a framework for moving forward. Both are neuroscientists at Caltech. Adolf's work focuses on humans, especially FMRI studies, and Anderson studies animals, including mice and flies. Thus they bring a very important perspective to the conversation. While some researchers argue that animals lack emotions, their stance is that not only do animals have emotions, but the study of emotions in animals is essential to understanding emotions in humans. By the end of this episode you will understand why they take this position. The reason they focus on creating a new framework is because it's not easy for scientists working in very different areas to share their work with one another. They hope to create a framework that will work across disciplines and make it easier for different disciplines to cross pollinate. In their preface they explain that their position is that emotions are real and that they occur in both humans and non human animals. They suggest objective criteria that can be used to determine which animals have emotions. The basic premise is emotions are states of the brain and the book is dedicated to Jacques Panksep. They start out with a critique of the movie Inside out because it's full of a lot of popular ideas that are wrong. So I'm going to go over these ideas and explain how they do not fit our current understanding despite their popularity. So the ideas that appear in the movie are that there are a few primary emotions, that emotions are triggered by specific external stimuli. Emotions control our behavior. Different emotions are located in different discrete brain areas. And the emotions are these conscious little homunculi, you know, the little people, cartoon characters in the movie. So these ideas are based on several assumptions. One is that primary emotions are irreducible. Another is that primary emotions correspond to those for which we have English names. Another assumption is that emotions are like reflexes and that specific emotions cause fixed and specific behaviors. Specific emotions occur in specific brain regions and emotions are purely subjective experiences. You probably recognize many of these ideas as representative of what most people think is true about emotions. Unfortunately, these ideas range from unproven emotions are irreducible to completely false. Lets consider the idea of the primary or universal emotions. Whether or not they themselves have smaller building blocks is an open scientific question. But there is growing evidence that emotions are influenced by culture and learning. So they're not always the same. And we actually don't know what is truly cross cultural or universal and what is shared across species. What about the idea that emotions are reflexes? Well, no, emotions are actually very different from reflexes because they are influenced by learning and they're actually flexible rather than fixed. Now, emotions do influence our behavior, but they don't control our behavior. In fact, both cognition and behavior are controlled by complex interrelated processes. Dr. Luis Pessoa, who I interviewed back in episode 106, has done a lot of work showing that the amygdala is involved in decision making, coming to the primary conclusion that emotions and cognition, which, when you define cognition as decision making, they are so intertwined as to really not be separable. That's why in his book he talks about the emotional cognitive brain. Now what about reflexes? Again, emotions may have evolved from reflexes, but they are, as I said, more complex and more flexible. And their evolutionary role appears to be promoting complex, flexible behavior. Discrete emotions do not come from discrete brain regions. There are no macroscopic areas or regions dedicated to specific emotions. Fear is not in the amygdala and I'll talk more about that in a few minutes. One important point that I will probably repeat is that the emotions as internal functional states are not the same as the conscious experience of emotions or feelings. An analogy to this is the various parts of the brain that participate in vision. If you look at them, you can't find vision in any of these regions. And there's not any region that has the experience of seeing feeling. And emotion is a whole person phenomenon. So the challenge is to create a framework that can be used across disciplines and across levels of analysis. And I'm not going to get into the details of that, except to emphasize the importance of providing a functional account, which means the focus is on what emotions do instead of the specific implementation or even the specific emotion. This also includes asking why emotions evolved. The tentative answer is to provide a flexible mechanism for dealing with environmental challenges. So you might think of the range as going from reflexes to emotions, all the way up to cognition or thinking, and all of them are needed for surviving in the world. So I don't want to focus on the technical aspects of this book, even though they're very important. But I want to take a moment to reflect on one of the key differences between the way neurobiology and cognitive neuroscience approach emotion and everything else for that matter. Systems neurobiology uses vocabulary from neurophysiology, cellular biology, chemistry, molecular biology. They pretty much avoid talking about the mind because their focus is the brain at the microscopic level. So that's neurobiology. Cognitive neuroscience uses a lot of terms from psychology, and they talk about the mind a lot because their focus is macroscopic. This creates difficulties in translating between the disciplines. There's nothing about science that requires that one field's explanations will map directly into another's. Some people think that psychology will be replaced by neuroscience, and other people think psychology offers a level of explanation that doesn't need to evoke neuroscience. My take on it is that I don't think neuroscience is going to replace psychology. But I do think it's important that psychology incorporate neuroscience to the extent that they make sure that their theories are consistent with neuroscience as we know it in any particular moment, so that they don't have a theory that is contradictory from neuroscience. Now, this dichotomy between the idea of psychology and other explanations, it's not the same as the classical materialist versus non materialist view of the mind versus the brain, because they're both materialist systems, but they're operating at different scales, macroscopic versus microscopic. The problem is, bridging from one scale to the other is not straightforward. One reason for this is that some properties are emergent and they just can't be predicted from the level below. A simple example of this is that wetness doesn't exist at the level of single molecules. So when we hear about large scale networks, small worlds and hubs, none of that easily maps down to lower levels of description. And really we don't know what the best level is for understanding emotions. The goal of neuroscience is to have a complete causal description that allows prediction. This is probably going to involve multiple levels. One subject that is discussed in great detail in this book Is how to get from correlation to causation. That's a very important topic that I'm going to basically skip today for time constraints. But one key point is lots of time we have correlation without being able to prove causation, Especially if we're dealing with things like FMRI data from human studies. That's one reason why we need the animal studies as a way to test whether correlations are causation and also to give us clues about where to look when we do human studies. So we're going to talk a little bit about what we have learned from animals, and I'm going to focus for a minute on what we have learned about the amygdala, because this is something that really has come especially from rodent studies, both the old ideas and newer ideas. First off, it's not the center of fear. Even if the amygdala is active during fear, that doesn't actually tell us much about the mechanism. A good analogy is you're looking at the speedometer of your car, and it tells you how fast you're going, but it doesn't tell you anything about how your car works. So how do we know that the amygdala is not the center of fear? Well, for one thing, when you destroy it, it doesn't necessarily impair all the aspects of fear. Some of that is from human studies. And as I said, we don't know its causal role. It might actually be part of a distributed process as far as fear goes. And we also don't know how the level of fear is encoded, what causes an animal to go from freezing to the next level of flight. Another thing that they found is with newer techniques in animals, they've found that the ventral medial hypothalamus contains neurons that are both necessary and sufficient for the expression of innate defensive behaviors, which is what they're now saying that they're studying in animals when they do these studies that used to be called fear. And it looks like the hypothalamus is actually also involved in coordinating central emotional states. And that's consistent with the fact that certain fear states can be generated in humans that don't have an amygdala. So be on the lookout for more discoveries that have to do with the role of the hypothalamus in emotion. This makes sense, given that the hypothalamus has a major role in homeostasis and controls things like appetite. And sometimes we wonder about things like, well, would hunger count as an emotion? And the usual answer is no, but it may just be that it's more of a continuum, going from drives like hunger to what we normally call emotion. The key take home point here is that the state of fear is not created by a single locus in the brain. It involves a series of distributed and interconnected nodes, which may actually change depending on the circumstances. So initially, the ideas about the amygdala and fear came from animal studies, and our more sophisticated ideas about the amygdala are also coming from animal studies. So what about invertebrates? Like insects, it turns out that they can show behavior very similar to that seen in rodents. With Drosophila, the fruit fly, they do experiments that look very similar, and they teach them to avoid, basically to do aversive Pavlovian conditioning. It exists in flies, and they're capable of associative learning, aside from doing experiments and demonstrating experiments where their behavior is surprisingly similar. Even though, remember that the fly has a much simpler brain, it's organized very differently, it has no amygdala and no hypothalamus. Studying the flies also resulted in some surprising discoveries about dopamine, which pop psychology has labeled as the reward chemical. In flies, dopamine is actually an aversive neurotransmitter. And then there are more recent studies that show that it could do both. And there are studies in rodents and primates that show that dopamine is involved in salience and arousal. What it does depends on which neuronal population is involved. So it's not the identity of the neurotransmitter, but the connectivity of the neurons that releases the neurotransmitter that matters. Also, they used to think that it was all about volume transmission, I.e. that dopamine was released kind of like a sprinkler and just hit everything in its vicinity. Whereas now it looks like it's much more specialized to specific synapses. They've also found subpopulations of cells in the ventromedial area that was traditionally associated with a reward that can actually be either reward or punishment. So dopamine does not encode valence, that is, whether it's good or bad. So the brain can actually flexibly attach a positive or negative valence to a previously neutral stimuli. This is important for learning. Of course, there's got to be some sort of hardwired encoding of valence, and this is probably represented by certain classes of dopaminergic neurons. So are we looking at emotional primitives? When we study flies, the argument for yes is that the aversive stimuli effectively trigger escape and avoidance behaviors across species. Flies, mice, other mammals, and flies also demonstrate the same sort of integrative effect, which is like the longer they're exposed to the stimulus, the more agitated they become, and their behavioral responses change based on how many times they're exposed. Their response persists for tenths of a seconds to minutes. So it's not a reflex. This is consistent with the idea of an internal state, but it doesn't prove that the internal state is. Is the same across species. Remember, as I said before, that the fly brain is organized differently from a mammal brain, and it looks like something called the mushroom body might be the analog of the hypothalamus. Now, the neurons are pretty similar, and they use the same chemicals. Of course, we know from what Dr. Seth Grant told us last month that there's probably some significant differences in the synapses. We don't have that many circuits characterized in the fly. In fact, there's only one so far. It's the giant fiber that goes from the central brain to the thoracic ganglia. Thoracic ganglia sort of acts like the spinal cord. So that's the circuit that mediates the reflex jump away from a looming threat. And so far, we don't know what else it does. Interestingly, there's a certain kind of restless behavior in flies where they walk on walls in a certain way that can be reduced by giving them anxiolytic drugs. So an interesting question presented in the book was, do flies have sex because it's rewarding or just because they're hardwired to do so? If you want to know the answer, you have to read the book. But the key idea here is that there are tools that can be used in invertebrates, like optogenetics, for example, that allow a level of resolution that we can't have in humans. And there's other animals like C. Elegans. It's unclear whether that animal has an emotion. But one thing we do know is we've got its wiring diagram, which certainly gives us a different level of resolution. So what happens when we move to studying emotions in humans? Well, obviously, the tools change dramatically. We rely on things like verbal reports and functional magnetic resonance imaging. FMRI problems with FMRI include its lower resolution. We're not at the level of the neurons because a voxel, which is a 1 millimeter square, will contain 10,000 to 100,000 neurons. Also, the emotions that are elicited in human experiments are much weaker, mostly because of ethical considerations. Also, language is involved in the human experiments, and this is both a strength and a weakness. You could argue that we aren't even studying emotions themselves because we're actually usually studying feelings, that is the conscious experience of the emotion, or studying the attribution of emotions to others, that is thinking. Besides fmri, some electrical stimulation experiments are done in humans. The problem is you can't really control where very well. However, we do have some broad conclusions based on electrical stimulation. One is that you can evoke conscious experiences of emotions by stimulating different cortical and subcortical regions. And there is some specificity and there do seem to be clear categories. Now, transcranial magnetic stimulation, which is a non invasive method, also has poor resolution, but it also has led to a few broad conclusions. One is that individual emotions can be compromised in relative isolation, and the other is that there is some anatomical segregation for processing individual emotions. Emotions. Basically what this is saying is that when you use tms, you are able to knock out emotions separately from one another. I want to say a few more things about functional mri, because in the past this is something that I have been very skeptical of and I had William Utah on and replayed his episode last year. So I should address this issue. These authors describe the problems with early FMRI and they also explain how the problems are being addressed, including the development of increasingly sophisticated statistical tools. So they argue that FMRI is much better than it used to be and is continuing to improve. They have statistical techniques for dealing with large amounts of data. They are getting higher quality data studies like the Human Connectome Project. So it is becoming possible to derive some causal insights. And there's a lot more detail about how this works in the book. Two other key points. One is that the majority of the historical data is of problematic quality. However, the future is bright as these techniques are rapidly evolving. Now, they do describe some human experiments that involve listening to music and watching movies. Researchers who can People who are watching movies are actually able to predict what they're feeling, but it's unclear exactly what it is they're measuring. It's back to that whole distinction between correlation and causation. The speedometer is a good predictor, even though it has no causative role. So they could be measuring correlation without causation. Their studies do seem to show widespread activation, which is why they need these, you know, real sophisticated techniques to detect patterns. In fact, that's what it's all about. Now With FMRI is finding patterns rather than asking, did that place light up? The key idea here is that these studies can show neural correlates without revealing causes. And actually, they argue that that's the biggest reason why we need the animal studies, because the. The animal studies right now are the only way of testing whether something might be a cause, because it's the only way to test at the level at which this could actually be determined. As I mentioned, the authors of this book did not propose any particular theory of their own. In fact, they discuss several theories, even though they argue that theories are premature. From my point of view, a good theory would be one that makes predictions. So let's consider a sample of the theories in the book. And actually, I would say that most of these theories are more about feeling emotions rather than primary emotions. There are the interoceptive theories, which date back to William James, who thought that all of our feelings of emotion were the direct result of interpreting interceptive signals from the body. Now, Bud Craig, who was on the show a while back, has spent his entire career studying the insula, and he has argued that that's the seat of emotional feelings and that only humans are capable of emotions, based on his definition. On the other hand, Antonio Damasio has argued that our feelings of emotion are basic to our sense of self and that they're an important element of consciousness. And more recently, he has acknowledged the importance of subcortical structures. Now, the constructed theory of emotion is also discussed. This is the theory that we introduced in episode 136, when I interviewed Dr. Lisa Feldman Barrett, who wrote the book How Emotions are Made. The authors of this book acknowledged that Feldman Barrett's theory is probably the most ambitious and comprehensive theory, though they criticize it for not making a clear separation between feelings and the other components. The thing that sets the constructive theory of emotion apart from the others is that it argues that the brain constructs emotional experience the same way that it constructs our other experiences. This appeals to me because it includes the role of prediction. And it's well established that perception is influenced by memory and expectation. This is true for vision, even though it's occurring outside of our conscious awareness or control. To me, it makes sense that a similar principle would apply to emotion, rather than emotion being some sort of outlier. Barrett challenges the idea that there are universal or primary emotions, arguing that emotions are learned in the sense that we learn to label our affective experience from the people around us. While the authors agree that learning and context affect emotions, they contend that emotions can be seen As a natural kind, they make the comparison to scientific categories like galaxies, planets, atoms, and molecules. They prefer Panksip's theory of the seven basic emotional systems based on subcortical circuits, though they acknowledge that his theory is based on animal research and Barrett's is based on human data. There are several other theories that I can't mention here, but I will briefly mention Joseph LeDoux's theories because they are quite antithetical to both Damasio and Panksip. LeDoux is now arguing that his work with rats was not about fear at all and that he has been studying what he calls the survival circuits. According to his view, animals might have conscious feelings when they're in a state of fear, but they don't know that they're afraid. He also rejects any evidence that emotions have subcortical origins. The authors Write on page 229. We emphatically believe that subcortical structures are necessary for the conscious experience of emotion, even though they may not be sufficient. LeDoux claims that only humans can experience emotion, a conclusion I think definitely outstrips the available evidence. It's one thing to say that we haven't proven that animals have emotions, but there is also no convincing evidence that they don't. From an evolutionary point of view, it seems more likely that emotions evolve before the higher cognitive functions that we associate with being human. Let's return for a moment to the difference between Panksip and Barrett, because I see value in both approaches. I find myself trying to reconcile the two. I find myself wondering if this is another example of poor translation between levels. Let me explain what I mean. If you look at the constructive theory of emotion, it's really a theory about feeling emotion, the conscious experience of emotion, which is one level up from the emotion state. So when Barrett argues that emotions are learned, she's not talking about basic affective experience, which likely includes a subcortical component. It's unclear that we will ever be able to study emotion states in humans, because we're almost always focused on the conscious experience of the people being studied. In fact, many times we are even another step removed, because if you ask somebody to identify the emotion of a person in a picture, that's about attribution. And one can argue that that tells us more about social communication than emotion. At this point, your reaction might be, so what? One reason we'd like to understand emotion better is because millions of people suffer from disorders like anxiety and depression. Our current treatments are mostly based on chance discoveries. But a better understanding of emotion could literally lead to more effective treatments. And I'm not just talking about drugs. One thing that Dr. Barrett discovered is that people who can describe their emotional feelings with more granularity actually have better mental health. This is an example of a practical consequence of the constructed theory of emotion. Instead of seeing emotions as something outside our control, it suggests that we can take actions that will influence our future emotional experience. As a physician, I see this as an important step forward. At the same time, I hope more research will be done on the subcortical origins of emotions and we will continue to gain a better appreciation for the emotional lives of non human animals. The Neuroscience of A New Synthesis by Ralph Adolph and David J. Anderson is a book that I would recommend for students and researchers in the field. It's really not a book for general readers. Starting from the basic premise that emotions are brain states, they argue that the evolutionary role of emotions is to allow for flexible responses to environmental challenges. Given that emotions require more complex nervous systems than reflexes, it's not unreasonable to see them as a step on the road to the so called v volitional control of behavior. The authors of the book argue for developing a framework that could apply across various levels and sub disciplines. They don't provide a theory of emotion, but they do argue, based on the best available evidence, that non human animals have emotions because we can do experiments at the molecular and cellular level only in animals. They argue that it is essential that a neuroscience of emotions cross species. They also provide a brief overview of our current knowledge, mostly focusing on debunking the most popular myths that are still circulating. So I want to end by reviewing a few basic principles. 1. The existence of primary emotions has not been established. 2. Emotions are not reflexes, they are affected by learning and allow for flexible behavior. Number three Emotions do not control our behavior. Number four you don't have little characters in your head controlling you like in inside out. 5. No single brain area is devoted to a specific emotion and that includes the idea that the amygdala is not the center of fear. Lastly, your experience of emotions is influenced by your culture and your life experience. I want to encourage you to check out the show notes for this episode@brainsciencepodcast.com where you'll also find episode transcripts. The show notes include links to previous episodes on the subject of emotion and links to all of our social media websites as well as the chance to subscribe to our free newsletter so you can get show notes automatically. That's brainsciencepodcast.com and you can send me email@brainsciencepodcastmail.com Now I do want to share just a few announcements. First, I want to remind you again about the trip to Australia in May of 2019. Please do write to me@brainsciencepodcastmail.com if you're interested and I will send you the details, including how much it costs. There's still plenty of time to sign up. I need to take just a minute to tell you about how you can support Brain Science. There is premium content available to those of you who subscribe to the premium subscription and also Patreon. The difference between them is that if you have premium, you get the entire back catalog as well as all episode transcripts. The back catalog now contains about 100 episodes. Both premium and Patreon supporters get new episode transcripts, ad free episodes and the audio for each month's Facebook Live. Next month's Facebook Live will be on the first Thursday of December at 8pm Central Time. To learn more about how to support Brain Science, just go to brainsciencepodcast.com donations even if you can't support the show financially, you can help out by telling other people about the show, subscribing and leaving a review in your favorite podcasting place. Apple Podcasts, Google Podcasts, Stitcher, Spotify, wherever you happen to listen. Now I'm going to take just a moment to tell you about my other podcast. First, for years I've had a show called Books and Ideas, but admittedly it has not had any episodes in 2018. To fix that, I will be posting an episode next month with Dr. Pamela Gay from Astronomy Cast. Pamela was recently inducted into the Academy of Podcasters hall of Fame and It's been about 10 years since she was on the show, so I'm looking forward to talking with her again. My goal is to begin to post books and ideas on a regular basis again in 2019. Traditionally, books and Ideas has been the place I put stuff that doesn't fit on this show. The big news that I want to share with you today is that I am starting a brand new podcast. In fact, I have started a brand new podcast. By the time you hear this episode, there'll be at least two episodes of the new show out. The new show is called Graying Rainbows coming out LGBT later in life. So you can find that just by searching for Graying Rainbows. I hope that you will share that show with anyone you think might be interested in it. We also have a Facebook page and that's a weekly show that comes out on Mondays. Graying rainbows coming out LGBT plus later in life Brain Science with Dr. Ginger Campbell is copyright 2018 to Virginia Campbell, MD. You can copy this show to share it with others, but for any other uses or derivatives, please contact me at brainsciencepodcastmail. Do it.