SUMMARY
Jaak Panskepp, who is an affective neuroscientist teamed up with ___, a clinician, to compile this book as a discussion of the foundational systems that our emotions are built upon. His view, based mainly on animal research, is that there exist core systems in the brain that structure our feelings and motivations, which he labels seeking, rage, fear, lust, care, panic/grief, and play. He argues that all of these systems are subcortical and evolutionarily ancient (in other words, that these basic emotions are not at all unique to humans), but that some more complex emotional states that are unique to humans come from an interplay between these core systems and cortex. Practically, the authors argue that understanding these core systems would help inform the treatment of emotional disorders in a way that is more useful than some of the therapy methods that existed at the time of writing.
The first chapter - ancestral passions - lays out the groundwork for the rest of the book. Panskepp first clearly outlines the fact the named systems that he will discuss may differ somewhat from folk psychology terms, which is why he's capitalizing them.
He also introduces the triangulation approach - by studying, 1) subjective states in humans (as related to the brain), 2) brain functions themselves, most easily studied in animals, and 3) behaviors that animals exhibit in order to survive (as related to the brain). The chapter also introduces the competing ideas of the basic or dimensional view of emotion. The basic view holding that there are different emotion systems, and the dimensional view postulating that a valence and arousal dimension are sufficient to explain most emotional content, subserved by a "core affect" brain process. He also introduces competing views about the separation of affect and cognition, which is his view, vs. a more unified view of affective neuroscience as part of cognition.
In the second chapter, "The Evolution of Affective Consciousness," the authors give a pretty comprehensive history of studying affects in other animals and map out where they believe this study has gone awry. Much of the chapter is devoted to arguing against the James-Lange theory of emotion and their modern versions, which is that evaluating internal sensations lead to the affect. In some versions of this theory, that requires language, a neocortex, working memory, or autonomic changes in other parts of the body, so Panskepp and Biven take some time to provide evidence against all those ideas. The more ancient (further back) the circuit for stimulation, the more intense the emotional display / affective feeling. Decorticated animals are highly emotional. People without nerve input from the rest of the body still experience emotions.
Then Panskepp begins delineating all of the primary process emotional systems in the brain. Chapter 3 is devoted to the SEEKING system, which most neuroscientists refer to as the reward system. This seeking / motivation system is foundational for many other emotional systems because it promotes behavioral activation (Panskepp thinks through affects of excitement and pleasure). This system runs up from the ventral tegmental area to three main destinations through distinct dopamine pathways (the medial forebrain bundle and later hypothalamus, the nucleus accumbens, and the medial prefrontal cortex. The system responds particularly to novelty and unexpected rewarding stimuli.
Chapter 4 is about the "rageful furies of the mind," which readily manifest when our SEEKING state is thwarted. The system depends on the periaqueductal gray, hypothalamus, and amygdala, although only stimulating the PAG in isolation can create anger. It is a very complex system that interacts with many neurotransmitters and higher level brain areas, and is difficult to completely distinguish from other violent behaviors such as predation, infanticide, social dominance, although it does seem to be distinct.
Chapter 5 is about fear. The fear system is a two-way pathway that goes from the central amygdala to the anterior/medial hypothalamus, and down to the periaqueductal gray in the midbrain (which was a very similar system as rage). It is also co-accompanied by physiological arousal like rage. The unconditioned stimuli of fear include pain, noise, predators or evidence of predators, open spaces, sudden movements. A number of neuropeptides can cause fear behaviors in animals.
Chatper 6 deviates a bit from the primary process affective systems, and makes a case for the affects in learning and memory. Here, the authors try to distinguish learning and memory from affects themselves, to again argue against the need for learning and the cortex to explain core affective systems. They do show though, that learning and memory are commonly associated with emotional arousal. For example, fear generates strong automatic and involuntary responses, and the seeking system is intimately involved in motivation to do the repetitive practice that leads to procedural skills.
Chapter 7 discusses the LUST system. The authors make the case that although the basic circuitry is the same for males and females, there are important differences in this system between sexes. In males, the center of sexual urges comes from the medial regions of the anterior hypothalamus. I found it interesting that testosterone activates this region (the preoptic area in rats) and that they will behave as if injections of testosterone to this area are pleasurable. The ventromedial hypothalamus is more influential in female sexual responsivity. They respond to different sensory input channels and promote different copulatory reflexes, but the evidence for lustful affects is still a bit uncertain. Humans appear to be a bit unique among mammals in that adrenal testosterone is important for female sexual enthusiasm in a way that is not important for other species. Vasopressin and oxytocin are also important neurochemical signalling molecules that seem to play different roles in males and females. Vasopressin makes males "pushy" but makes females sexually inhibited. Oxytocin is more abundant in females and encourages nurturing behaviors, although this also may differ in humans.
Chapter 8 discusses the CARE system. This is the first of a "super-cluster" of three primary processes that are behind non-sexual social bonds (including grief and play). Each of these systems shares similar regulatory neurochemistries, and are a little bit more contentious in terms of other scientists thinking of them as primary-process systems. The authors think of care as developing as part of parental instincts (and particularly maternal instincts) towards rearing young. They posit that the care system developed evolutionarily from the older LUST system (because the neuropeptides involved in mammals and birds are similar to those found in non-caring reptilian LUST systems). The system seems to rely on the paraventricular nucleus (PVN) and dorsal preoptic area of the anterior hypothalamus, which are cell fields that produce oxytocin in mammals. Destroying the cells there also remove maternal behavior. There is also a pathway from this system to the ventral tegmental area, which activates seeking behaviors. Although the system exists in both females and males, it is highly sensitized and affected by female sex hormones, and can change during pregnancy and lactation. Endogenous opiods are also strongly involved in this system, and the authors posit that opioid addiction and withdrawal are very similar and rely on the same neural systems as social dependence.
Chapter 9 discusses the panic grief system, which is the other side of the coin from the CARE system. This chapter about separation distress is really personal to Panskepp as he discusses his profound grief to the death of his daughter Tiina. This system involves the anterior cingulate, the dorsomedial thalamus, the periaqueductal gray, and cerebellum. In animals, some smaller nuclei in the hypothalamus and midbrain have also been identified including the ventral septal area, dorsal preoptic area, and bed nucleus of the stria terminalis. This grief system seems to oppose the care system, in that the care system can be activated in relief of separation. It probably evolved from a more ancient pain network in order to encourage maternal attachments and then social bonds. It gradually becomes more insensitive as animals age.
In chapter 10, "playful dreamlike circuits of the brain," the authors discuss social joy and laughter. Playfulness is fragile, in that it only occurs when an animal is safe, secure, and feeling good. It is distinct from both seeking and aggression. Touch stimuli can initiate play which are processed in distinct areas of the thalamus (the parafascicular complex and posterior dorsomedial thalamic nuclei). The system is also tuned to unpredictable stimuli. Laughter seems to be involved in the system, and all the sites you can stimulate laughter (via the SEEKING system), animals will also self-stimulate.
Chapter 11 addresses the question of the subjective self more generally. The authors believe that subjective affects can motivate behavior only within the context of also feeling like you have a self. They believe this sense is generated specifically by deep subcortical processes. This is built on top of a "mapping" circuit that integrates emotional, motivational, homeostatic, and perhaps sensation (a protoself). They postulate that visual/auditory stimuli actually began as affective experiences rather than as sensory experiences. Enhanced sensory/perceptual systems are built on top of those. The authors don't go too much detail in discussing details of a core-self network, but generally localize it to the subcortical midline structures (midbrain, hypothalamus, thalamus, and parts of the cerebellum). Which are most important for conscious experience? Not the cerebellum or basal ganglia, but the rest of the structures are candidates, and he doesn't really elaborate too much which ones. I think that's because he's providing evidence mostly from lesion studies, and a lot of lesions to the midbrain, hypothalamus, and thalamus destroy consciousness, wakefulness, and dramatically alter behavior.
Chapter 12 tries to integrate the brain science with psychotherapeutic ramifications. The core theme of this chapter is that directly manipulating the primary-process emotional circuits (through psychology, somatic, or physiological approaches) will add to manipulating tertiary-emotional processes (through mediation of cognitive processes that is at the core of cognitive-behavioral therapy and much psychotherapy). For example, engaging opposing positive affective systems to counter-act negative disorders, or replaying negative memories in the context of care and play. The authors also strongly advocate neurochemical manipulation of these systems.
Finally, there is a conclusion where the authors reiterate many of the major philosophical themes of the book, including the need and importance of affective research generally, the idea that affect and consciousness is evolutionarily old and that other animals are likely to experience very similar primary-affects as us (especially mammals), and the need to distinguish between the primary-process affective networks mediated by the midbrain to the tertiary-process networks that involve the interaction between cognitive systems and emotion.
-----------------------------------------------------DISCUSSION QUESTIONS----------------------------------------------------
Are affects intentional (always about something)? Payel and I have noticed that when she experiences an emotion, she can almost always immediately know the cause. My experience is that emotions are much more clouded. I often find myself developing competing hypotheses about what is making me feel a certain way. Where do you fall on this scale?
The triangulation approach (in parallel looking looking at the mammalian brain, the instinctual emotional behaviors of other animals, and the subjective states of the human mind) strikes me as reasonable if we are reasoning by analogy. However, it also seems circular to say things like "it provides a way of understanding scientifically, for the first time, some of the experiences of other animals." I feel like the inherent assumption is that subjective experience is created globally in the brain rather than some specific module supported to it. That also seems reasonable, but I wish that Panskepp spent more time discussing the assumptions behind making inferences from such an approach. What did you think of the triangulation approach?
Do some animals (perhaps those that reproduce quickly and are predated on) mainly exist in a constant state of negative affects?
The distinction between operational terms "reinforcements, rewards, punishments" and affective terms "satisfaction, discomfort" is interesting to think about from the perspective of artificial intelligence. We know how to structure algorithms to reach some goal state or avoid some undesirable state. We can pre-define those states for a variety of types of systems. Do you think this is inherently different than what biology does in emotional systems? If so do you believe that even simple artificial intelligence experiences affective states? If not, how do you justify why you believe animals (including humans) feel affects (as Panskepp does), but artificial intelligence does not?
Do you agree that the "play" system represents a core primary-process affect? This was an interesting section, but also seemed the weakest to me, in terms of describing a core affect. To me, it seems like "feeling good" is a more primary affect that accompanies play.
It is interesting that Panskepp defines primary process emotion systems based on mostly neural (rather than affective) criteria. In some ways, this seems like a really good thing to do. Cognitive psychologists are constantly arguing about "what attention is" for instance, but writing from a mechanistic perspective clarifies things when our language has gotten a concept wrong. However, we should probably pay attention to when something is defined according to one view, but doesn't feel intuitively true. For example, I feel "fear" very obviously in my experience, but I don't feel "play" very obviously in the same way. Instead I would say that I feel "happy" while playing.
- Unconditioned stimuli can activate the emotion
- Unconditioned behavioral / autonomic changes occur to the unconditioned stimulus
- Gates and valuates incoming stimuli
- Positive feedback outlasts the original stimulus
- Can be regulated by higher cognitive functions
- Influence higher mental processes
- Generates affective feelings
Naming "red" is not required to see "red." But the situation is more complex than Panskepp is revealing. For instance, can people who frequently name more colors experience finer distinctions between them? This perceptual expertise could also apply to affects.
The intense pleasure and excitement of stimulating the "seeking" system, really seems to undermine the goal in some forms of Buddhism to eradicate suffering through eliminating craving. It seems like this approach cannot work without also eliminating the joy of stimulating this "craving" or seeking system. I was curious what the Buddhists in the group made of how much animals enjoy stimulating this system.
Is it correct to draw distinctions between affects and other perceptions? How do we distinguish things like hunger, thirst, and pain from the affective systems that Panskepp and Biven delineate?
One thing that was very striking to me throughout is that very small structures in the midbrain or hypothalamus play a huge role in emotion (and thus in learning, subjective experience, and behavior). Yet, these areas are often just left out of the discussion in cognitive functional imaging experiments in humans because the areas is so small. We have the capacity to see some of these areas now with advances in ultra-high field imaging, and this book has been personally inspiring to me to refocus on this anatomy and neglected regions.
I especially liked predictions from this book about different "flavors" of similar affects. For instance, there is anxiety rooted in low level activation of the fear system, and anxiety rooted in the separation anxiety of the GRIEF system. These likely come from different sources, and may be treated differently. Were there other surprising distinctions that you noticed?
This view of focusing on neural systems pre-empted the research domain criteria (RDOC) that the National Institute of Mental Health uses to investigate mental disorders rather than the diagnoses in the Diagnostic and Statistical Manual of Mental Disorders. He advocates for emotional endophenotypes instead of syndromal thinking in psychiatry. It was interesting to me that RDOC ended up creating domains around arousal and valence, rather than the core affective systems that Panskepp advocates for. Could we make even more progress by rearranging research domains in terms of neural systems rather than symptoms?
I'm curious how the statements Panskepp makes about the amount of neuroscience research devoted to affects vs. to vision and hearing are surprising to non-neuroscientist readers. It seems to me that much more popular neuroscience is devoted to emotion and motivation than to perception. Yet, the statement seems true to me when I try to think about the detail of neural mechanisms supporting these functions. For example, an expert in vision might be able to describe the basic structure of the system and advise a computer science about how to recreate a similar system. Hearing, touch, and motor action are a little bit further behind, but getting closer. For emotion, an expert could tell you the regions that seemed to be involved in what part of the process and the chemical transmitters or modulators that seem important, but not even approach the mechanistic detail that a vision scientist could give. Is this surprising? What do you think caused this to happen; is it a bias that animals do not experience emotion in the same way as humans, and so cannot be used as good mechanistic models? Is it that we lack intuition about what is actually being represented and computed by the system (which is very intuitive in sensory domains). Do you have other ideas?
I enjoyed thinking about the distinction between tertiary (reflective cognitive) and primary process emotions as a way to seek out higher level intelligence in other animals. For instance, humans seek out things that make them afraid when they are in a safe context (like horror movies or haunted houses). We might infer that animals that seek out feeling afraid in safe contexts are also reflective. Or animals who avoid a pleasurable emotion because of longer-term consequences. Can you think of other examples of these types of tests?