Deric Bownds

Chapter 10

Emotional Mind

The rambling internal narrative of our thoughts is like a swimmer in a sea of emotions, constantly registering the gentle rocking that goes with subtle changes in background feelings as well as the crashing waves that accompany fear or passion. What are our emotions? They certainly feel more rich, juicy, and energetic than "just thinking." We experience them as coordinated responses spreading through the whole body, frequently linked to a social context. Sometimes we may experience a sense of "losing it"---getting so caught up that rational processes aren't running the show. Could this be because a different brain has taken over at such moments, one adapted to our evolutionary past but not necessarily to our present world? Does this mean that emotion and reason are opposing forces?

An emerging consensus argues the contrary position--that our rational processes depend vitally on the "lower" brain structures that mediate emotion, and that lack of emotion can be just as damaging to reason as its excess. The apparatus of rationality is not layered on top of our basic regulatory and emotional processes, but rather rises and develops from within them. "I feel, therefore I think" reflects what happens in the brain better than Descartes's famous "I think, therefore I am." 1 Modern clinical studies on human patients with brain lesions show that either an absence or an excess of emotions can prevent rational behavior. We might begin to think of emotions as the glue that binds together the modules of mind and as pointers to the contents of consciousness that are relevant to a particular moment. Emotions are a part of our evolved psychology and are every bit as cognitive as the more rarefied domains of mathematical and linguistic logic. This concept has been overlooked because emotions are so much harder to describe. There also has been a cultural bias, with intellectuals setting up an artificial opposition between reason and emotion in which the emotional sides of our natures are unstable and unreliable distractions from "real thought." Fortunately, the long-standing scientific taboo against studying subjective experience is being relaxed, and more weight is being given to affective neuroscience alongside, or as a part of, cognitive neuroscience. 2 It is interesting in this regard that computer scientists who model cognitive processes are sometimes finding it useful to introduce global analogs of affect into their simulations.

This chapter begins with a brief discussion of the way we define emotions and then moves on to consider both their evolution and the brain structures and pathways that underlie our emotional behaviors. Lesions in some of these structures not only diminish affect but also distort rational processes, which suggests that emotional pathways must be intact for our rational faculties to work properly. Emotional pathways can work more rapidly than those involved in conscious reasoning, and this may give us some insight into why we don't always act the way hindsight tells us we should have. It is easy to misapply our emotional repertoire. We all face the problem of applying hardwired emotional machinery such as the fight-or-flight response to modern circumstances in which threats are much more subtle than a charging lion on the African savanna.

Defining Emotions

Devising a reasonable system to describe the many emotional behaviors we experience is a daunting task, but this hasn't kept numerous different schemes from being launched. This examination of fear in the accompanying self-experiment, along with many other emotional situations, suggests that our emotional experiences have at least three different manifestations: (1) a physiological component that includes autonomic nervous system activity---changes in viscera, blood pressure, distribution of blood flow, digestive system, and so on; (2) our behaviors, such as facial expressions, that convey anger or sadness; and (3) our subjective feelings, such as love, fear, and hate. Some of our emotions reflect drives or desires basic to survival, such as hunger, thirst, lust, pain, pleasure, and aggression. These emotions, in some form, are shared by all higher vertebrates, especially mammals. Other emotions are more explicitly related to communication with our fellow humans: happiness, disgust, surprise, sadness, anger, distress, interest, fear, and jealousy. Measuring emotions is a challenge, because the same stimuli can be pleasant or unpleasant, depending on the context. For example, if you have just had a huge candy bar, you may react negatively to a sweet soft drink, but if you are tired and thirsty, the same sugary drink is a pleasing relief.



A useful place to begin a description of the variety of emotions is with our own experience. Let's take the example of fear. Pull yourself away from this focused reading for a moment and try to recall receiving a fright: perhaps you narrowly escaped being hit by a speeding car or became aware that someone was following you on a dark and dangerous street. Your heart pounded, your body tensed, your face contorted with fear, and a strong feeling of dread passed through your body. Probably you can actually experience a faint "replay" of these reactions just by recalling the situation.


It is interesting but not surprising that most studies on brains and emotions deal with negative affect, such as fear, anxiety, and anger. We don't go to doctors to seek cures for feeling happy, joyful, or optimistic. 3 It is easier to elicit and measure fear or anxiety than happiness, and much of what we experience as positive emotions might be mainly the absence of negative ones. Further, happiness is linked not simply to the absence of a threat, but also to how we perceive our general well-being with respect to that of others.

We can't say which emotions are "primary" in the same sense that we know the primary colors are blue, yellow, and red---we can't even be sure the analogy holds. But the main candidates include anger, sadness, fear, enjoyment, love, surprise, disgust, and shame. 4 Each of these emotions is made up of a family of related feelings. Anger, for example, might be expressed as fury, outrage, resentment, exasperation, acrimony, irritability, or hostility. Enjoyment includes relief, contentment, delight, amusement, sensual pleasure, rapture, euphoria, and (at the extreme) mania. An intense emotion usually does not last more than a few minutes, but we can experience its more muted form, mood, for many hours. Finally, each of us has a distinctive temperament---a readiness to evoke certain sets of emotions, such as those associated with being cheery or melancholy. Most of these temperaments remain remarkably constant throughout life. There is considerable evidence that some temperaments, such as novelty seeking, optimism, and pessimism, are moderately heritable and not greatly influenced by family environment. 5 Beyond these dispositions are the more blatant disorders of emotions such as chronic anxiety or depression. Categorization of all these states is complex, and researchers disagree about some very fundamental issues:

How many basic universal humans emotions are there?

Should emotions be described as discrete entities, such as anger, fear, and disgust, or as points on continua, such as pleasant-unpleasant, aroused-unaroused, approach-avoid? Is there really a difference between the two descriptions?

What is an unambiguous test of whether an emotion has occurred? 6

Emotions are hard to define not only behaviorally but also anatomically, because many different brain regions come into play in every emotion. 7 Emotional circuits include the limbic system, which forms a ring around the brainstem and consists of cortex around the corpus callosum (mainly in the cingulate area) as well as cortex on the medial surface of the temporal lobe, including the hippocampus and amygdala (refer to Figures 3-3 and 3-4). 8 These regions communicate between higher centers in the prefrontal and association cortexes and the hypothalamus. The hypothalamus contains centers whose stimulation can elicit stereotyped emotional performances. The location of these structures is shown in Figure 10-1. At the present time, we don't have a theory of the anatomical circuits of emotions, as we do for vision. The study of emotion, just like the study of visual or other kinds of cognition, requires a dissection of emotional processes into elementary operations. Such an effort is still in its early stages.

Figure 10-1

Brain structures that are important in generating emotions. We are looking at the brain from the left side, as we have before in describing the surface features of the cortex, but now the entire left side of the brain has been removed so that we have a medial view of the inside surface of the exposed right hemisphere. Adapted from Figure 16.4 in Bear et al., Neuroscience : Exploring the Brain.

Emotions Are Evolutionary Adaptations

The psychologist Nicholas Humphrey, whose ideas were mentioned in Chapter 2, suggests that we might trace the origins of affect back to such simple behaviors as the chemotaxis of Escherichia coli or the wriggles of acceptance or rejection displayed by amoebae. In more complex animals, specialized receptors evolved to report useful or noxious stimuli to the brain. Early in the evolution of the brain, sensory and emotional sensitivity was always linked to a corresponding bodily activity. A subsequent development was to make the responses a brain activity (attraction or repulsion) that was felt but not necessarily acted on (recall the itch that you can choose not to scratch.) The suggestion is that to "like" a stimulus might be to respond to it in such a way as to keep up or increase the stimulation (positive feedback), and to "dislike" it would be to respond in such a way as to keep down or reduce it (negative feedback). Affect might be linked with the way a stimulus is evaluated at a submodal level. For example, among visual stimuli, red light is typically exciting whereas blue is calming. Among the different kinds of tactile sensations, itches are irritating, gentle stroking pleasurable. Within the gustatory modality sweet tastes are pleasant, rotten tastes revolting. 9



The origin of emotions might be traced back to responses, in simple animals, of "liking" or "not liking" some aspect of their environment. Emotional, or affective, behaviors probably evolved because they conferred some selective advantage on those who had them.


Moving through vertebrate evolution toward humans, we can trace an increase in the complexity of autonomic nervous systems and emotional behaviors. One view is that the emotions are an evolutionary by-product of the neural regulation of the autonomic nervous system. Recall that a basic job of the sympathetic and parasympathetic divisions of the autonomic system is regulating the body during mobilization and energy expenditure, as well as during rest and recovery. During mobilization, the sympathetic system releases the neurotransmitter norepinephrine at many end organs---for example, on muscle to increase the speed and effectiveness of its contractions. Parasympathetic activity is approximately the mirror image of this, a calming and return to emphasis on vegetative self-maintenance and restorative functions. It uses the neurotransmitter acetylcholine and is active during sleep. These systems are established in bony fishes and become more complex in amphibians and reptiles. In mammals a new complex of nerves appears, the ventral vagal system, which innervates the heart and other visceral organs, as well as facial and vocal muscles involved in emotional expression. This is a system thought to permit emotional mobilization short of the full sympathetic arousal of the fight-or-flight response. 10 It has components that inhibit sympathetic arousal to promote bonding and affiliative behavior. Emotional circuits appear to be much more malleable than hardwired reflexes such as pain withdrawal and the knee-jerk response. There can be large variations between individuals and between related groups of animals. Think of the difference between Staffordshire terriers (pit bulls) and St. Bernards or between common and bonobo chimpanzees (described in Chapter 4). These different variants of the same species have evolved very different emotional behaviors.

Emotions and the Physical Environment

We obtain some interesting clues about the possible origins of our emotions by examining common human phobias, which are resistant to change but can be extinguished by conditioning therapy. It is striking that there are only about two dozen common elicitors of phobic reactions, such as insects, snake shapes, heights, looming large objects, and growling noises, all of which would have been very relevant to the survival of our ancestors. We appear to have an innate predisposition to associate aversive stimuli with these objects. Psychological experiments have shown that most of us look at a picture of a boa constrictor without getting too rattled, but if that picture is paired with an electric shock just once, the next time we see the snake picture we become quite agitated. If we are shown a picture of a flower together with the shock, it takes a much longer series of pairings of picture and shock to induce the same level of aversion to the picture. More extensive experiments have been done in animals to demonstrate innate predispositions to certain kinds of conditioning. There can be single-trial learning of a response to selective stimuli even if there is a long delay between stimulus and result, and the learning is not easily extinguished. Laboratory rats become averse to bright light and noise if these are linked with pain but not if they are linked with nausea, whereas a particular taste can become aversive if linked with nausea but not if linked with bright light or noise. 11



Learning that is obviously relevant to survival, such as becoming averse to the taste of some food that once made you ill, is much easier and more rapid than learning to be repelled by a picture of a flower with which a shock is associated. Selective pressures over evolutionary time have favored formation of the brain circuits underlying such predispositions.


There is general consensus that our limbic system is the primary site of innate or preorganized primary emotions, containing circuitry that is especially responsive to features of the sort just listed. 12 An unresolved issue is to what extent something like a "snake detector" might be hardwired by genetic instructions. We can sketch a plausible scenario for how genes might specify a detector for the "feature set" of a long, thin object that moves sinuously. However, it seems implausible that they might code the precise coloring or shape of the animal. Different parts of the world have snakes of different shapes and sizes. Similarly, the face detector cells mentioned in Chapter 8 and below might actually start as more generic icon detectors for certain sets of curves and lines, as in the "happy face" image. This image is made only of a circle containing an upwardly curved arc in its bottom half, positioned under one dot for the nose and two for the eyes. Newborn human children recognize this icon and distinguish it from others in which the elements have been scrambled.

Emotions and Social Exchange

Charles Darwin recognized the importance of emotions as signals that accurately inform one animal about the state of another. 13 This is most clear in higher mammals like ourselves. All of us think that we can tell when a dog or cat is happy, playful, angry, fearful, curious, anxious, or in some other state in which an adjective ordinarily descriptive of human emotions seems appropriate. We know for ourselves, and usually assume it to be true for animals, that emotions are messages about our own internal state as well as direct, nonverbal signals to others. Here we might go with our common sense, as not inappropriately anthropomorphic, for some similar underlying mechanisms are clearly at work in both ourselves and other mammals. 14

Emotions can inform our intelligence what to think about, and in this sense we are, to use a phrase attributed to Richard Dawkins, "walking archives of ancestral wisdom." Practitioners of evolutionary psychology argue that common emotional states---fear of predators, guilt, sexual jealousy, rage, grief, and so on---are adaptations to repetitive features of the environment faced by the small groups of hunter-gatherers in Africa who are thought to be the ancestors of all modern humans. 15 For example, discovering one's mate having sex with someone else signals a situation that threatens future reproduction and present "investment allocation." This cue should therefore activate sexual jealousy. The discussion of evolutionary psychology in Chapter 5 mentioned the importance of detecting cheating, and many aspects of anger, guilt, and shame revolve around this issue.



Basic emotional states common to humans of all cultures, such as enjoyment, sadness, anger, disgust, and fear, correlate with biologically significant events in the world, marking them as good or bad, as associated with life or death. These emotions, and their counterparts in other animals, are adaptations that have the purpose of informing or influencing the behavior of others.


The psychologist Steven Pinker suggests that our intellect is designed to relinquish control to our passions so that they serve as reliable guarantors of promises or threats that allay suspicions of double-crosses or bluffing. This would explain why we unconsciously advertise emotions on our faces---why our being "unable to control our emotions" is a guarantor of their authenticity. For the most part, it seems a good design to segregate voluntary cognitive systems from mainly involuntary housekeeping systems that regulate blood pressure, skin glands, and so on. But to prove an emotion's authenticity, perhaps selection has bound that emotion to a physiological control circuit (such as quavering, trembling, croaking, weeping, blushing, blanching, flushing, or sweating) over which we normally don't have much voluntary control. If we really want to know whether people are trying to fool us, or are displaying sham emotions, we insist on doing business with them in the flesh, so that we can see what makes them sweat. These arguments from evolutionary psychology, of course, contend that our emotions are behaviors that maximize not our fitness today but that of our ancestors in a now-vanished past. In this view, we humans are "adaptation executors" who respond to present circumstances in ways developed in our evolutionary past.

Subcortical Systems Underlying Emotions

Emotions were identified with lower brain centers in the 19th century through experiments on dogs whose cerebral cortexes had been removed, leaving the diencephalon intact (including the thalamus and hypothalamus). These animals could still show rage and fear. In the 1930s, Papez proposed that the limbic system, acting through the hypothalamus, was the anatomical basis of emotions. During this period, brain lesion studies showed striking changes (such as loss of fear, indiscriminate eating, and increased sexual activity) in the emotional behavior of monkeys whose temporal lobes had been removed. Lesions in the frontal cortex, paralimbic cortex, or amygdala caused reduced and inappropriate social interaction. A similar syndrome (Kluver-Bucy syndrome) is found in humans with bilateral damage to the amygdala and inferior temporal cortex. The discovery that removing the frontal lobes of aggressive monkeys calmed them led to psychosurgery, an outstanding example being the frontal lobotomy of humans. This procedure not only calms the subjects but also severely impairs their social and affective behavior.

The Autonomic Background of Emotions

Before focusing on the brain, however, let's start further downstairs, with the peripheral sympathetic and parasympathetic nervous systems that were introduced in Chapter 3 and mentioned above. These systems mediate our bodily experiences of emotions. Recall that our sympathetic nervous system is engaged during arousal and emergencies, expending energy, whereas the parasympathetic system regulates rest, recovery, and energy storage. The ratio of parasympathetic to sympathetic activation, set by the hypothalamus in the brain, is a central component of our affective life, correlating with whether we are relaxed, open, and engaging, versus avoiding, defensive, or aggressive.



You can try a simple exercise to sense some of the global body correlates that go with activation of the peripheral sympathetic and parasympathetic nervous systems. The following exercise is probably best done by having someone read it out loud to you, but you can also read the instructions and try it on yourself. Close your eyes for a moment and imagine yourself on vacation, sitting on a rock by the ocean watching a beautiful sunset. Let time slow down so that all you are aware of is the colors in the sky and the gentle breeze from the sea. Be completely in the present moment. Let the pleasant feelings in your body take you back to an earlier time, perhaps when you were a child, when you felt secure and cared for. After you have spent at least a few minutes getting into this fantasy, just note how your body feels, note your breathing, and note what your shoulder and jaw muscles are doing. Now try a different fantasy, one less pleasant. Imagine yourself to be walking, late at night, down a deserted city street with only one or two streetlights. There are no other people around. You suddenly hear a scuffling noise, and four rough-looking men tumble from an alley 30 feet in front you. This surprises you and you stiffen a bit, as though you don't want them to notice you. But then, the moment they see you, they abruptly turn to face you, fan out, and start to approach you, slowly and deliberately, with very flat expressions on their faces. You look for a place to run, but it's too late. They break into a run towards you. Now scan your body again, and note how it feels. Even if you just read through these descriptions without really getting into them, you can probably sense differences in your body---in your breathing or in how tightly you are holding your muscles.


The self-experiment in this section has the goal of eliciting body changes that go with parasympathetic (relaxing) versus sympathetic (arousal) activation---changes that are intense enough for you to sense them clearly. Consider, now, that much of your daily attitude or temperament might be described as being somewhere between these two extremes. Indeed, the relative activation of these two systems sets a somatic background, or bias, that can influence your behaviors.

Strong sympathetic or parasympathetic activations can set a tone or mood that persists for some time. Think of the times, for example, when you have experienced a persistent sympathetic activation as you remained a bit jumpy for many minutes after being excited by suddenly having to avoid some danger, perhaps a speeding car. Norepinephrine released by your sympathetic nerves was causing increased attentiveness and excitation that decayed only slowly as the norepinephrine was taken back up into nerve cells.

Brainstem Modulation of Attention and Appetite

Against this backdrop of overall sympathetic or parasympathetic activation, a whole host of mood-altering chemicals regulate our appetitive and attentional behaviors. These global modulators of our emotions are manufactured in restricted regions of our brainstem or basal forebrain and then sent via long axons to widespread regions of the cortex, where they are released. Central players are neurotransmitters such as the biogenic amines (dopamine, norepinephrine, and serotonin) and acetylcholine. These systems serve as brain "spritzers" regulating our states of alertness, anxiety, elation, depression, or aggression. 16 They can speed up or slow down the rate at which our mental operations proceed.

Figure 10-2 shows the location of one of these systems, a small group of cells called the locus coeruleus. It is essentially an outpost of the sympathetic nervous system that can extend that system's arousal into the brain. Axons from these cells travel to multiple locations in the cerebral cortex and cerebellar cortex to release norepinephrine, which enhances our attention and memory processes. Another spritzing system releases the neurotransmitter serotonin in the brain. Reductions in the activity of circuits that use norepinephrine and serotonin apparently contribute to depression in many people. Serotonin levels appear to be inversely correlated with aggression. For example, mice that lack a certain postsynaptic receptor for serotonin are more aggressive in stressful situations. Lowering serotonin by drugs or genetic manipulation increases aggressiveness, and raising it has the opposite effect. 17 Low levels of brain serotonin are associated with anxiety in humans and also in animal models for anxiety, and inhibitors of serotonin uptake into nerve cells - which raise the concentration of serotonin outside the cells - are used as antidepressants. The psychoactive drug LSD mimics the action of serotonin.

Figure 10-2

A diffuse modulatory circuit originates in the locus coeruleus. This brain nucleus, which occurs on both the left and the right sides of the pons regions of the brain, has only about 12,000 neurons. A single one of these neurons can make more than 250,000 connections, sending axon branches to both the cerebral cortex and the cerebellum. These release norepinephrine, which can enhance arousal, attention, and memory.

Dopamine, whose role in movement we mentioned briefly in Chapter 9, acts as a central activator. If nerve terminals secreting it are destroyed in rats, the animals lose desire, motivation, adaptability, and exploratory behavior. Dopamine levels are affected by cocaine and amphetamine, and they play a role in responses to heroin, alcohol, nicotine, and cannabis. 18 Chemicals that modify dopamine receptors can alter drug-seeking behaviors. 19 Some people who show stronger than average novelty-seeking behavior have a variant of one of the dopamine receptors. 20

Dopamine-containing cells in the midbrain are activated by novel appetitive stimuli, but not by aversive stimuli. 21 This observation has suggested the simple model that when something interesting or good happens, these cells in our midbrain shower higher brain centers with dopamine, which causes an appetitive behavior. 22 Cocaine and amphetamines mimic this effect by inhibiting the reuptake of dopamine into presynaptic terminals, thus maintaining activation of synapses longer than normal. Studies using mice initially showed that when the transporter that normally removes dopamine from synapses was deleted, the animals became hyperactive and also indifferent to cocaine and amphetamine. They presumably already had the "high" that the drugs usually cause. (These experiments are done by constructing "knockout" mutants in which the gene for the transporter has been deleted. 23 ) Alas, the clarity suggested by the initial knockout experiments did not persist, because repetition of this and similar experiments found that knockout mice in which either the dopamine or the serotonin uptake mechanisms were deleted were indeed sensitive to cocaine. 24 Recent work has now suggested that brain circuits using the neurotransmitter glutamic acid are involved in producing the changes that lead to compulsive drug seeking.



Several different centers in the brainstem contain nerve cell bodies that send axons to many different areas of the cortex and cerebellum. Their activation can cause global changes in attention, alertness, appetite, and motivation.


Further experiments have suggested that the purpose of the dopamine system is not to produce feeling of pleasure but to draw attention to events that predict rewards, so that the animal learns, recognizes, and then repeats them. 25 Dopamine surges are measured in an area called the nucleus accumbens during anticipation of food or sex. This could explain why many drugs that stimulate the dopamine system can drive continued use without producing pleasure, as when cocaine addicts continue to take hits long after the euphoric effects of the drug have worn off or when smokers still smoke after cigarettes become distasteful. These drugs have essentially hijacked a brain mechanism that evolved to signal rewards relevant to fitness and reproduction (such as food or sex) and instead enslave the organism to desires that have no adaptive value. Brain imaging using PET shows activation in the mesolimbic dopamine system as addicts describe feelings of intense craving for cocaine. (The mesolimbic dopamine system connects the orbitofrontal cortex, in the prefrontal area behind the forehead, with the amygdala and nucleus accumbens.) Different areas are preferentially activated during rush (the drug euphoria, or high) and during craving for the drug.

Other Correlations of Chemistry and Emotional Behavior

We can list several further correlations of chemistry and emotional behavior. Morphine and other opiates can block pain pathways in the brain, sometimes causing euphoria, by mimicking a different class of neurotransmitters made from peptides (small pieces of proteins). The brain contains complex families of these peptides, including a group referred to as the opioid peptides: the enkephalins, endorphins, and dynorphins. They also play a complex role in regulating social attachments and the distress caused by social isolation. 26 Another peptide, oxytocin, appears to play a role in soothing and bonding behaviors, eliciting the parasympathetic opposite of the sympathetic fight-or-flight response. 27 In some rodents (different species of prairie voles), monogamous versus polygamous social structure and behavior correlate with different expressions of the oxytocin and vasopressin systems. Some chemicals that influence arousal, anxiety, or aggression can come from sources outside the brain. Aggression (especially aggression between males) is influenced by testosterone, as we noted in Chapter 6. Castration of male mice reduces their aggressive behavior, and injections of testosterone can then reinstate it. 28 Epinephrine and adrenocorticosteroid levels correlate with arousal and excitability; they typically are higher in emotive than in repressed individuals. 29

All of these systems certainly are regulated by our conscious cognitive response to objective aspects of our environment, especially whether it is nurturing or threatening. There are other internal explanations for why we feel this or that particular way. Many of the autonomic and brainstem systems discussed in the previous sections have cycles of their own, alternating periods of activity with intervals of rest and renewal of their neurotransmitter systems. This has been well documented in the cyclical activation of cholinergic versus adrenergic systems in sleep versus wakefulness, which is described in Chapter 12, and in changes in cognition and mood that accompany the monthly ovulation cycle in women. Perhaps we should consider that these cycles can generate feelings for which "there is no reason" in the usual sense, not withstanding our tendency always to invent a reason for the way we are feeling. Experiments with hypnotized, split-brain, or anosognosic patients have shown that we will cheerfully confabulate bogus reasons for a behavior if the real cause is not accessible to our consciousness. We should consider the possibility that during some feelings and emotions, the brainstem systems are calling the shots, and our fertile imaginations are merely supplying a cover-up story.



Thinking about the relationship between our brain chemistry and our emotional moods and thoughts presents us with the "Which came first, the chicken or the egg?" problem. Cognitive processes can direct the mood-altering chemistry that originates in the brainstem, and spontaneous changes in this chemistry can alter our thoughts. In the latter case, there may be no "reason" for our thoughts or feelings, in the way we commonly suppose.


Chemical imbalances in these systems can underlie affective disorders. We are learning that distinctive chemistries correlate with defense or arousal, with rest, with anticipation of novelty or appetitive behavior, with anxiety or depression, and so on. Legions of biotechnologists currently employed by pharmaceutical companies are trying to synthesize compounds that selectively enhance or depress these systems and provide more effective therapies for depression, anxiety, drug addiction, and anhedonia (the inability to experience pleasure).

Higher Levels of Emotional Mind

Continuing to trace the components of our affective life further up into the brain, we encounter the limbic system that was mentioned in Chapter 3 (see the section "Layers of the Brain"). It is thought to mediate the basic survival-related programs of the brainstem (such as feeding, aggression, and sexuality) and also to regulate the social emotions of distress, bonding, and nurturing. The limbic structures shown in Figure 10-1 communicate between higher cortical centers and the hypothalamus, also mentioned in Chapter 3 as a central control center for coordinating our emotional responses. Electrical stimulation of nuclei of the lateral hypothalamus, in sequence from front to back, elicits appetitive behaviors related to temperature regulation, sex, drink, and food. Stimulation of medial structures in the hypothalamus seems to produce opposite, aversive behaviors. The hypothalamus directs the autonomic nervous system and integrates and coordinates the rapid behavioral expression of emotional states. Lesions or electrode stimulation can generate discrete, stereotyped expressions of fear, anger, pleasure, or contentment. Each of these emotions seems to possess the animal completely during its expression. You may have heard about a "pleasure center" in the brain. Actually, several such centers have been identified in rat and human brains.

Finally, the top element of the triune brain is the cerebral cortex overlaying the limbic system. Its frontal lobes are central to the regulation of emotions. Mood disorders have been correlated with abnormalities in the ventral prefrontal cortex. 30 This region is consistently underactivated in familial depressive patients and is activated during periods of mania. Activity in our prefrontal lobes may be part of the answer to a fundamental question: How do we assemble the vast array of feelings we have in different situations, avoiding some and desiring others?

Emotional Mind as a Foundation of Rational Mind

The neuropsychologist Antonio Damasio 31 has suggested that that during our development we accumulate a large number of learned or acquired secondary emotions that necessitate activity of the prefrontal cortex; in this they differ from the more innate primary emotions associated with the limbic system. These secondary emotions are essentially feelings learned by associating primary emotions with different situations we encounter while growing up---situations that mark them as positive or negative, to be approached or avoided. A small positive or negative stamp is put on almost everything we encounter. We develop a set of rules, a morality, for what to do and what not to do. This training, the acquiring of secondary emotions, involves our visceral or autonomic nervous system. When we have aversive or positive reactions to something, our autonomic nervous system responds, usually outside of our normal consciousness, by changing such variables as the rate of moisture release by our skin, our heart rate, and our blood pressure. (Changes in skin conductance can be used as a convenient and crude assay of this autonomic nervous system arousal and have been found to correlate with changes in activation of portions of the cingulate and motor cortex.) Subtle changes in muscle tension over the body may also accompany the pairings of feelings and autonomic activation. Damasio uses the term somatic markers to refer to these body correlates of feelings. These markers are discussed below, in the section "Emotional Responses Can Be More Rapid Than Reasoned Ones."



You can get a sense of the processes involved in acquiring secondary emotions if you pause for a moment, close your eyes, and recall a physical place---a house or room, perhaps---where you had either a very happy or a very unhappy experience, perhaps in your childhood. If, after a period of years, you recall this space, return to it, or perhaps enter a space that reminds you of it, what do you feel? Can you sense subtle, possibly flickering, changes in your body, perhaps small relaxings or tensings of muscles, that accompany the feelings?


The existence of learning processes by which we accumulate secondary emotions are revealed by their disruption in patients who have lesions or tumors in their frontal lobes, particularly the right ventro-medial part. Their behavior is superficially normal most of the time---intellect and memory seem intact---but motivation, foresight, goal formation, and decision making are flawed. They are subject to bursts of immodest, impolite, or other inappropriate behaviors; they act almost as though their free will had been taken from them. Their emotions and feeling also are diminished.

Normal individuals, as well as those with brain damage located outside the frontal lobes, generate skin responses that indicate autonomic arousal when they are shown disturbing or erotically charged pictures. Patients with frontal lobe damage fail to generate these responses, in spite of realizing that the content of the pictures ought to be disturbing. A typical response might be "I know I ought to react to that, but I can't. I just don't feel anything." It is fascinating that primary emotions (such as reacting fearfully to an angry face or an aggressive growl) can still be intact, but emotions based on experience, cognition, and learning are impaired. It is as though the link between frontal cortex and limbic system, the link to feeling in the body, has been severed. The frontal lobe patients appear no longer to be able to connect logic with good or bad. They do endless cost-benefit analyses of simple issues, such as when to schedule an appointment with the doctor, without being able to reach a decision. Their choices do not appear to be informed by their emotional histories. The use of their intelligence in the pursuit of a goal (deciding when to schedule that appointment) has been compromised. This observation suggests that emotions may be important in the selection of goals one at a time, so that they don't conflict.

A consequence of being unable to accumulate a feeling of the worth of different strategies during learning is a myopia for the future. In a clever gambling experiment in which cards are drawn from different decks for money or fines, normal patients begin to employ an advantageous strategy before they actually realize what strategy works best. They also generate an anticipatory skin conductance response if they ponder a choice that turns out to be risky, before they know explicitly that it is risky. Frontally damaged patients show no anticipatory responses to risky choices and continue to choose disadvantageously even after they know the correct strategy. 32 Patients with large lesions elsewhere in the brain can learn to succeed in the gambling game just as normal subjects do, even if they are aphasic and indicate that they can make no sense of what is going on. Imaging experiments have now shown increased activity in several frontal cortical regions that correlates with learning reinforced by rewards. The implication of this finding is that reasons, at least some of them, require emotions.



Our thinking recruits our bodies. We don't plan for or anticipate futures unless we can link them with affect---with memories and hunches from similar situations accumulated over our personal histories.


Lateral Organization of Emotions

Reason and emotions are contrasted in our folk psychology, common-sense talk, and experience, and indeed they can correspond to activities in different parts of the brain. Damage to the right temporal area that is homologous to Wernicke's area in the left hemisphere leads to disturbance in comprehending the emotional content of language, whereas damage to the right frontal area homologous to Broca's area in the left leads to difficulty in expressing emotional aspects of language. Disorders of affective language, called aprosodias, can be classified as sensory, motor, and conduction aprosodias in the same way that the aphasias that compromise syntax and grammar are classified. Their anatomical organization mirrors the organization in the left hemisphere. Thus patients with aphasia can show not only cognitive defects in language but also defects in the affective components of language: intonation of speech (prosody) and emotional gesturing. PET scans show that when a subject is instructed to comprehend the meaning of words spoken in a foreign language, blood flow to the left hemisphere increases, whereas flow to the right hemisphere increases when the subject is asked to evaluate the emotional tone of the same spoken passage. 33 These patterns of localization that form during development of a typical brain are not completely predetermined, because, as we noted in Chapter 6, young children in whom the left cerebral hemisphere is severely damaged or even removed early in life can develop an essentially normal range of language functions.

A number of experiments show that the left visual field (which projects to the left hemisphere) is superior at correctly identifying faces, whereas the right visual field (projecting to the right hemisphere) is better at perceiving facial expressions and emotions. The left ear (projecting to the right hemisphere) is better at detecting the emotional tones of voices, whereas the right is better at identifying their content. Lesions in the temporal lobe and temporal lobe epilepsy can cause a variety of emotional effects. Right temporal lobe lesions, as well as epileptic foci, not only can destroy ability to understand the affective content of language but also can trigger paranoia, anger, delusion, sexual feeling, déjà vu, and hallucinations.

During happy affect, the left hemisphere, particularly the frontal lobe, is more active than the right, and the opposite is true during unhappy affect. Brain lesions in the left frontal area are more likely to be associated with depression than similar lesions on the right. Work of the psychologist Richard Davidson has shown that strong positive feelings and approach behaviors correlate with left frontal lobe activation, whereas withdrawal and fear correlate with right frontal activation. 34 Imaging studies are now showing in some detail the subareas within each hemisphere whose activities correlate with appetitive, pleasant, or pleasurable emotions. They can be distinguished from other areas whose activation correlates with aversive, negative, or unpleasant emotions. Activation of the left frontal lobe correlates with suppression of activity in the amygdala (discussed below), which is central to fear responses. Facial behaviors of happiness or disgust also correlate with the left/right frontal activation pattern. Emotional content of sensory stimuli influences even the primary sensory areas of the brain, for pictures with strong positive or negative affect cause significantly greater activation of the visual cortical areas discussed in Chapter 8.



Stable individual differences in the relative activation of the two hemispheres show a correlation with a person's basic temperament, sense of well-being, and vulnerability to depression and bad feelings. For each of us there may be a genetically determined set-point for temperament, analogous to brain regulation of our body metabolism to maintain a preset weight. 35


Do you think there are differences in the expression of emotions on each side of the body? Why is the smile of the Mona Lisa perceived as ambiguous? She is smiling with only the left side of her mouth, the part controlled by the right hemisphere, which is more active during negative affect. 36 Which of the faces shown in Figure 10-3 do you perceive as happier or more pleasant? Most people say the face on the right looks happier. It has its right lip, controlled by the left hemisphere, curled upward. Facial expressions are not always symmetrical but rather tend to emphasize the left side of the face. 37 This reflects the fact that the right hemisphere, controlling the left side of the face, has a more pronounced influence than the left in controlling emotional expression. This is true in humans as well as in other primates. 38 Try to test lateralization of your own expressions by doing the accompanying self-experiment.

Figure 10-3

Mirror-image faces. Which of these faces looks happier to you?



You can experience lateralization of your emotions firsthand with a few simple facial expressions. Try curling up just the right side of your lips (controlled by left hemisphere). Does this feel sort of like a smile? Now, curl up just the left side of your lips (controlled by the right hemisphere). Most people report that it doesn't feel nearly as friendly.


Emotional Responses Can Be More Rapid Than Reasoned Ones

Why is it that we don't always do what we think we should do? Why do our emotions seem to have minds of their own? Recent findings have suggested an answer: Our emotions can use neural networks different from those used for conscious reasoned responses, and they can short-circuit those responses. 39 This may help to explain why we have so little reflective insight into our emotional life.

Emotions are special-purpose organizers that can perform rapid positive-negative evaluations of events below our consciousness; they are rapid reaction systems especially useful for dealing with encounters with other people. A person who resembles someone who once caused us anxiety can trigger an immediate emotional response before rational thought cuts in. Repeated exposure can decondition the anxiety. For example, subway commuters who continually face strangers start to like best those whom they have seen most frequently. Emotional responses seem to be front-line reactions that spring quickly into place before rational deliberation has had time to function. They reflect the read-out of an emotional memory that is the basis of preconscious filters of the sort described in Chapter 8 for the visual system. These filters can act on our perceptions to give a slight positive or negative affective bias even to stimuli that superficially appear neutral. The building of our emotional history might be described as the development of a large number of somatic markers: correlated and rapid reactions of the visceral, somatic muscular, and emotional brain pathways that have become linked to repetitive aspects of our environment, both positive and negative. 40 An example is the lasting influence of persons we admired or feared in our early childhood, causing us to react unconsciously (with attraction or withdrawal) to people who have features reminiscent of those people we knew decades ago. 41



Pause for a moment and repeat the word "lamujuva" to yourself several times. Now, do the same with the word "rakachaka." Does one of these words seeming more pleasant than the other? (Most English-speakers like the first and dislike the second.) 42 This suggests that the brain tags these percepts with a value, even though they carry no explicit cognitive meaning. Indeed, the brain may assign a value to virtually every perception.


Central Role of the Amygdala

The amygdala and hippocampus are key structures in forming our emotional memories and in evaluating emotional stimuli. 43 Although the hippocampus has long been implicated in emotion, the current view is that it is more involved with cognition and learning than with emotion. The hippocampus would be involved in recognizing a face, perhaps that of a cousin. The amygdala might add that you really don't like him. The amygdala appears to be required for the acquisition of conditioned autonomic responses to visual or auditory stimuli, but in the absence of the hippocampus, it is not sufficient for recalling the declarative knowledge relevant to those stimuli. The hippocampus, though required for the acquisition of such declarative knowledge, cannot support the autonomic conditioning in the absence of the amygdala. 44

The amygdala is a structure with an interesting evolutionary history. Its interactions with the thalamus play a major role in the mental life of fish, reptiles, and birds, whereas in mammals it is increasingly relegated to more specialized functions. Neurons of the amygdala show Pavlovian fear conditioning, and the whole structure plays a central role in triggering aversive and fear reactions. 45 Such reactions can be triggered outside of our normal awareness if a fear-inducing picture is flashed for a brief instant before a masking stimulus that we actually perceive. We report nothing unusual, even though changes in skin conductance that are characteristic of a fearful response are recorded. 46 Imaging studies show activation of the amygdala when human subjects encounter aversive stimuli, 47 during aversive conditioning, and also in masking experiments. The storage of unpleasant memories associated with aversive stimuli is apparently facilitated by sympathetic arousal, for when human subjects are given propranolol (a drug that blocks the action of adrenaline) during exposure to an unpleasant story with pictures, they are less likely to remember the story than subjects who receive a placebo drug.



The amygdala enhances the storage and persistence of emotional memories in a process that is facilitated by the adrenaline released during stress. 48


The amygdala sends projections to all areas of the cortex, regulating an emotional bias of cognitive functions. 49 In fact, there are many more nerve cells sending information from the amygdala to the cortex than from the cortex to the amygdala. This might be part of the reason why emotions can be so persistent and refractory to cognitive control. Its output is wired to parts of the brain needed to produce a panoply of fearful behaviors: shortness of breath, jumpiness, tension, and diarrhea. Different portions of the amygdala regulate different aspects of fear responses (such as autonomic nervous system arousal, emotional behaviors, and the hypothalamic-pituitary stress response). Each of these areas can be lesioned separately. 50 (There are so many different structural and functional units within what is usually called the amygdala that some argue that it is misleading to refer to "the amygdala" as a structural and functional unit.) There appears to be a negative correlation between left prefrontal lobe activation and activation in the amygdala, such that the relative suppression of left frontal activity that correlates with withdrawal and depression goes with increased amygdala activity. 51 Fearful situations that activate the amygdala can alter the intensity of conditioned responses. 52 This appears to be what happens when, in a tense or threatening situation, our reaction to an unexpected noise is much greater than it would normally be.



With a group of friends, you can conduct a simple demonstration of how activation of fear by amygdala-related mechanisms can influence our behavior. Start talking to the group about any old topic and then, without warning, startle your listeners by suddenly making a loud and unpleasant noise, such as slamming a book on a table. Then continue with your talking. After a while, begin to tell a story that becomes increasingly frightening. You might, for example, spin a tale of reading in the paper about a homicidal inmate escaping from a local prison and then later, while reading alone at night, hearing what sounded like someone trying to force an upstairs window open . . . etc. Now, again without warning, repeat the unpleasant noise you made previously and note whether the intensity of your friends' startle reaction has increased and whether they report this to have happened. (This exercise has to be done with some skill so that your friends don't figure out ahead of time what you are up to!)


We all seem to have a "negative-making or fear-making system" that can run semi-autonomously, the amygdala being a sort of "command central" that can program our overall emotional bias toward being apprehensive or anxious. We could view the arousal of this system, as happens during the period of heightened excitability that follows a sudden fright, essentially as "an amygdala attack." Its opponent would be a "positive-making system" that correlates with enhanced left frontal activity that can suppress the negative system and generate positive appetitive behavior. Test the idea of two such systems against your daily experience. Does the metaphor of "positive-making" or "negative-making" machines inside you, sometimes uncoupled from the input that may have initiated them but still biasing your reactions and decisions, make sense to you?

It is interesting that the amygdala of humans is more fully formed at birth than the hippocampus and develops more rapidly, such that fear behaviors are fully developed by 7--12 months. This may be why we remember essentially nothing from our first 2 years of life but apparently retain emotional experiences. It may also explain why these experiences, which psychoanalysts have long pointed to as a key to later emotional life, are so potent and so difficult to understand rationally. Behavioral and drug studies on monkeys have shown that corresponding fear behaviors appear at 9--12 weeks after birth, along with a peak of synapse formation in the prefrontal cortex and limbic system, including the amygdala. 53

Joseph LeDoux and his collaborators have done a series of fascinating experiments to demonstrate that the amygdala can trigger emotional reactions before the cortex has fully processed the triggering input. 54 Previous experiments had shown that damaging part of the amygdala can abolish learned fear responses. In one set of studies, rats were trained to fear a flashing light by having the light paired with a shock. In normal rats, the fear response is extinguished if the light is displayed regularly without the shock. This process takes several weeks. If the visual cortex is removed, rats still learn to fear the lights, but when shown the lights without the shock for several weeks, they retain fearfulness, unlike rats with intact cortexes. The idea is that once it is established, an emotional aversion is permanent, but the cortex can learn to dilute that aversion by inhibiting the amygdala's response. The pathways suggested by LeDoux's experiment are summarized in Figure 10-4.

Figure 10-4

A schematic drawing of relationships among the thalamus, amygdala, and cortex. Sensory input via the thalamus to the amygdala permits crude identification of potentially significant features of the environment and rapid reaction to them. More careful identification of the stimulus uses slower cortical circuits. Adapted from LeDoux' The Emotional Brain.

One possible explanation for post-traumatic stress disorder is that this cortical inhibition has been impaired. This model may explain how phobias are formed in humans and why they can be so tenacious. For a Vietnam veteran suffering from post-traumatic stress disorder, a clap of thunder may evoke the sweating and pounding heart he experienced during a battle. 55 In people who have unlearned a phobia, a single scary experience can sometimes bring it back in full force. This suggests that the phobia has not been lost to emotional memory, even though the behavior has been extinguished under most conditions.

The amygdala is just one synapse away from the thalamus, whereas the hippocampus is several additional synapses away. That means it takes a sensory signal as much as 40 milliseconds longer to reach the hippocampus than to reach the amygdala. That time gap allows the amygdala to respond to an alarming situation before the hippocampus does. When out of the corner of your eye you see what looks like a snake, it is the amygdala that sends the signal of alarm that makes you jump. You react even before the hippocampus has had time to coordinate figuring out, in consultation with other parts of the cortex, whether what you saw was actually a snake or a piece of rope. The amygdala appears to be a central part of the rapidly acting pathway that tags many words and images with positive or negative affect before they reach awareness.



Experiments done on rat brains reveal a neural basis for reacting to something emotionally faster than we can think. Nerve pathways through the amygdala that short-cut or bypass slower cortical circuits follow the "better to be safe than sorry" principle and presumably, given that they exist in all unimpaired present-day members of our species, made the difference between life and death for our ancestors.


Perception of Emotional and Cognitive Pathways

Because cognition and emotion emphasize different pathways, we might expect that their relative timing could be varied, with emotion occurring before or after thought. For example, we might recall a mental image of a snake and then, immediately afterward, the feelings that go with it, or we might notice a long, thin, sinuous object and have an aversive feeling before we note whether it is a snake or not. Try the accompanying self-experiment to examine the timing of your thoughts and emotions.



This simple exercise might permit you to experience dissociation of feelings and cognition in yourself. It would be best to ask a friend to read these instructions to you, but again, you should be able to do the exercise by yourself: Sit comfortably, close your eyes, breathe gently, and attend to the breathing. Notice, but do not hold onto, thoughts that pass through your mind. Let thoughts enter your mind as easily as they go, but always gently return to being aware of your breathing. Next, notice the thoughts that arise and are accompanied by positive or negative feelings. When do you notice the thought, perhaps an image of a person or social situation, and when do you notice the feeling that goes with it? Do they arrive separately or together?


The purpose of the self-experiment is for you to experience associating spontaneous thoughts or images with affect on a time scale of 100--300 milliseconds. This is the time scale of the flickering of thoughts and the somatic markers that go with them. You may have noticed that when an image spontaneously occurs in your mind, your awareness of the cognitive content of the image (such as the identity or visualization of a face, or the relationships of a social context) occurs slightly before you experience the emotion usually associated with that image. This fact is used in cognitive therapy and disciplines such as Buddhist mindfulness meditation to permit dissociation of an image from the affect normally associated with the image. It is during the first milliseconds of connecting images and affect (particularly the negative affect of feeling threatened, afraid, or powerless), that self-awareness might intervene to present other options. After this window has closed, negative affect usually becomes harder to dislodge as it reinforces itself through other associations.

If we calm down and pay quiet attention to what is going on in our heads, we can experience many fascinating phenomena that reveal how the brain works over tenth-of-a-second intervals. Chapter 12 describes several experiments that demonstrate that the brain can rearrange time and space in the period of 50--400 milliseconds after an external or internal visual image is initiated. This is also the time frame of the exercises on initiating motor patterns that we tried in Chapter 9.

Up to this point, we have mentioned an array of balancing systems---sympathetic and parasympathetic, brainstem spritzers, right frontal and left frontal plus amygdala activation---that all contribute to the formation of our background emotions or predispositions. To these systems, we must add the systems that mark images and ideas from our emotional histories with limbic, visceral, and muscular correlates to form an array of somatic markers. These markers are elicited as recurring situations are encountered, and they can act as little filters or machines inside our head---autopilots that sometimes can take a neutral event and put a positive or negative spin on it. Such largely unconscious mechanisms can attach small quantities of emotional energy to ongoing events, determining, when someone cancels a lunch with us, whether we think "she doesn't like me," or "he must be very busy," or determining, when we confront a challenge, whether our immediate reaction is "I can't do it" or "I can do it."

We are also endowed with an amazing kind of cognition, exercised in the self-experiment, that can gain some awareness of these lower processes going on. This awareness of thinking, or thinking about thinking, is sometimes called metacognition. It is a limited ability to sense and review the vast number of rapidly acting autopilots that have assembled during our development. We couldn't live without these automatic routines. We don't reinvent the wheel each time we encounter a familiar situation; rather, we reserve our higher cortical resources for dealing with what seems novel. Still, when the underground routines are not working, our higher faculties can be used to sense, tinker with, and possibly alter them.

Modern cognitive therapies represents a very practical application of this ability. We have already made reference, in Chapter 3 (in the section "Imaging the Activity of the Brain"), to patients with obsessive-compulsive disorder (the behaviors exhibited may include excessive hand washing or ritualistic counting and checking). Some have learned to control their behavior by thinking, when they feel the initial emotional urge, "That isn't me. That is a part of my brain that is not working." Or "That feeling is just the result of a false message from my brain." As the aberrant behavior diminishes, the same decreases in frontal lobe activity that are observed during effective drug therapy occur. What is being developed here is the literal equivalent of a mental immune system. Just as we acquire an immune history---knowledge of pathogens that have invaded our system and that now can be quickly expunged if they appear again---so we can acquire a history of maladaptive emotional routines. The analog of the macrophage or T cell is a trained recognition mechanism that has become very sensitive to the onset of these states and can trigger cognitive procedures that essentially excise them, as soon as they are detected, by activating more functional routines.



The use of cognitive therapy techniques to reduce inappropriate emotional behaviors and actually change brain activation patterns provides an example of how consciousness can play a role in regulating brain physiology.


Facial Musculature and the Communication of Emotions

Having delved in some detail into the internal mechanisms that regulate our emotional life, let's draw back to consider some social aspects of emotions and the central role played by faces and facial expressions. The amygdala is a key player in the interpretation of expressions that communicate emotions. It receives information from various areas of the visual cortex, particularly the portion of the inferior temporal cortex that contains face-responsive neurons. It also sends information back to all these areas---a possible means by which affective states could modulate sensory processing. 56 This could be part of the explanation for the biasing of perception by emotions that many psychological studies have revealed. 57 Subjects with bilateral amygdala damage are much less discerning in their judgment of other faces. Recent studies have shown that the amygdala can be activated by the masked presentation of emotional facial expressions. When a subject is shown a fearful or angry face for about 30 milliseconds, followed by a neutral face for 170 milliseconds, only the neutral face is remembered or reported, but brain imaging reveals an increase in activation of the right amygdala that is not observed when the 30-millisecond exposure is to a happy or neutral face. Interestingly, conscious aversive conditioning to an angry face paired with a burst of unpleasant noise correlates with activation of the left amygdala.

These systems originate far back in our vertebrate lineage. We don't hesitate much in interpreting the facial expressions that dogs and cats use to communicate anger, fear, or pleasure. The same can be said, with a bit less clarity, of many higher mammals. Nonhuman primates have facial expressions and meanings that are less subtle than our own but that seem obviously similar. We observe a dramatic increase in the complexity of facial musculature and expressions in moving from monkeys to apes to humans. Increasingly, the face is turned into a kind of semaphore, signaling complex social interactions. As we noted in Chapter 5, the use of facial and vocal muscles in communication was a component of the development of the mimetic intelligence that is thought to have provided the foundation for the appearance of modern language and mythic intelligence.



During hominid evolution, the increasing number and complexity of facial muscles supported the development of an array of emotional expressions that is universal across modern cultures. The amygdala plays a central role in both interpreting and orchestrating the response to these expressions.


We all know from our daily lives that our emotions correlate with, and are communicated by, a complex set of facial muscle signals. These signals are only one of several parallel layers of the kinesic communication (body language) that is occurring as we talk with others. This is presumably why more brain area is devoted to facial musculature than to any other surface of the body. These areas direct stereotyped combinations of muscle contractions to denote surprise, fear, disgust, happiness, and grief. 58 Darwin noted the universality of these patterns---how all people, from Oxford dons to Australian aborigines, express grief by contracting their facial muscles in the same way. 59 Flirting signals too are the same across cultures: a lowering of the eyelids or the head, followed by direct eye contact.

The most telling experiments involve the presentation of photographs of Western faces showing basic emotions to New Guinea natives who have lived for millennia in isolation from all other humans except their immediate tribe. Studies conducted on three such isolated primitive cultures show Western faces to Stone Age people, and vice versa, with the same results. 60 The same stereotyped sets of muscle contractions are used by all. The psychologist Paul Ekman has cataloged 80 different facial muscles used in communication. These muscles can be trained separately to reconstruct emotional expressions, and each expression can be specified as a list of relevant muscle contractions or relaxations. Some of the named muscle sets for different emotions are cross-cultural, others culture-specific.

What about the changes we feel in our bodies as our faces are smiling and relaxed, or fearful and tense? Fear is sometimes correlated with mobilization of the sympathetic nervous system; happiness is more likely to be associated with parasympathetic activation. Are the parts of our cortex that drive facial muscles also directing the autonomic correlates of the facial expression? Or is some sort of feedback from the facial muscles responsible for autonomic changes? Davidson and Ekman, mentioned above, have addressed this issue in an interesting experiment on smiling. 61 We all can recognize fake smiles, often used in deception, and we are fairly acute observers of the muscles involved. The difference is that the real smile that reflects happiness or joy involves not only the zygomatic major muscles which reach down from the cheekbones and attach to the corners of the lips, but also movement of the lateral part of the muscle (the orbicularis oculi) that encircles the eyes (making "crow's feet" at the corner of the eye). Electrical measurements show that when subjects are instructed to contract both of these muscles (but are not instructed to feel happy), the left frontal lobe becomes more active, just as in people who are experiencing positive emotions. This presumably accounts for the effectiveness of a form of therapy against depression: instructing the depressed person to move the muscles that make a smile. Doing so causes a lightening of underlying mood.

These are fundamentally important experiments, because they show that an emotional facial expression is not just a unidirectional command from the emotional centers of the brain to the face muscles, but rather that the facial expressions are part of a neural network in which activating one node (voluntarily activating the smiling muscles) can activate another node (the neural correlates of positive affect in the left frontal lobe). Recent experiments show that facial expressions of sadness, fear, and anger are similarly correlated with autonomic changes measured by skin conductance as well as with brain activation patterns measured by magnetic resonance imaging.



You can do a simple exercise on yourself to appreciate the relationship between facial musculature and emotions. Pause in your reading right now, slow down, and pay attention to your breathing for 20--30 seconds. Now, recall an experience that made you very happy, perhaps with a friend or lover---something you really enjoyed, maybe it made you laugh or smile; take a moment to get into that and try to relive the emotion you felt. Now, don't smile; make your face absolutely flat. What does this do to the feeling? Is it harder to hold on to? Can you feel a whole muscle set that goes with the internal feeling of emotion? Is it possible to separate them? Let's try just one more. This time, think of something that made or makes you angry, maybe another person you are angry at. Imagine confronting the situation or person. Let yourself feel really angry, maybe wanting to strike out. Now, stop contracting and tensing your face; let it go flat. What happens to the angry feeling? Again, can you feel the correlation? Now, try to smile and feel angry at the same time.


Misapplication of Ancestral Emotions---The Chronic Stress Response

What is the characteristic configuration of our autonomic nervous system and its correlated emotions? --for many of us living in modern industrial societies, it's not a very healthy one: chronic activation of a sympathetic nervous system designed for use only in episodic emergencies. This point deserves some elaboration. Each of us is blessed with an emotional repertoire controlling the four F's mentioned in Chapter 3 (fighting, fleeing, feeding, fornicating). These systems have evolved progressively through our vertebrate lineage up to our adaptations to the life of hunter-gatherers in the Paleolithic. The problem is that we frequently employ them in ways that are inappropriate in our contemporary surroundings. Many of us in complex high-tech and fast-paced societies chronically activate the stress response, rather than reserving it for genuine life-threatening emergencies. The endocrinologist Robert Sapolsky points out that we don't have the common sense of zebras, who don't get ulcers: They have a better handle on when to chill out (while grazing on grass) and when to go into high-gear stress (when the lion appears). 62 Our physiological response mechanisms are superbly adapted for short-term emergencies, but our bodies are not built to sustain their long-term activation. The things that work for short-term emergencies, getting fuel rapidly from storage sites and inhibiting further energy storage, are debilitating in the long run. Suppressing immune function in the short term, for example, allows the energy required to be used elsewhere, but in the long run, it makes us more susceptible to infections and cancer. 63

Our stress response recruits a diverse array of hormones, but it can be described in a simple way as occurring in two waves (see Figure 10-5). Within seconds of our encountering a stressor, cells in the hypothalamus cause activation of the sympathetic nervous system, and catecholamines (epinephrine and norepinephrine) are released by both sympathetic nerves and the adrenal gland. These act to mobilize the viscera and muscle for quick action. These amines also activate the amygdala, which is central in orchestrating the behavioral reactions to a stressful event, but their prolonged release in prefrontal cortex can cause cognitive dysfunction. As a further reaction to a stressful thought or event, on the time scale of minutes, our hypothalamus releases corticotropin-releasing factor. This triggers ACTH (corticotropin) release from the pituitary, and ACTH moves through the blood stream to the adrenal gland, where it triggers release of glucocorticoids that stimulate glucose metabolism and suppress immune function. (It is worth mentioning that there can be another, opposite kind of reaction to stress and challenges: Parasympathetic rather than sympathetic overactivation can lead to a "freezing" sort of reaction that slows the heart, shutting down the body rather than mobilizing it. Such a slowing down may underlie a striking phenomenon observed in some Stone Age cultures: Individuals who have been have cursed or ostracized by their group become increasing immobile and eventually die.)



Chronic sympathetic activation and glucocorticoid release have a large number of deleterious effects. A partial list includes immune suppression, atherosclerosis, digestive disorders, and accelerated aging and death of nerve cells in the hippocampus and temporal lobes. Stress also is known to cause decreases in levels of the neurotransmitter acetylcholine in the cortex and hippocampus, impairing learning and memory.


Figure 10-5

Stages of the stress response. Within seconds of the onset of a stressful event, norepinephrine and epinephrine (also called noradrenaline and adrenaline) are released by sympathetic nerve endings and the interior portion of the adrenal gland (the adrenal medulla). These chemicals enhance the readiness and excitability of both nerve and muscle tissue. More slowly, ACTH is released into the blood stream by the pituitary gland and carried to the adrenal cortex, where it triggers the release of glucocorticoids that stimulate metabolism.

The nervous, endocrine, and immune systems form a complex network that shares hormone and neurotransmitter molecules. 64 Interactions within this network are the focus of the new field of psychoneuroimmunology. A class of peptides used as neurotransmitters is also used in signaling by cells of the immune system. Nerve cell endings in the skin can secrete chemicals that shut down the immune system nearby. 65 This may partly explain why diseases such as psoriasis are exacerbated by stress. Prolonged exposure to stress hormones causes death of immune system cells and atrophy of neuronal connections in the hippocampus, leading to decay in memory and cognitive functions. 66 Prolonged depression is correlated with decreases in the volume of the hippocampus and the amygdala. 67 The molecular mechanisms by which glucocorticoids can block immune system activation are being unraveled. 68 Stress also decreases production of a trophic factor called brain-derived neurotrophic factor (BDNF), which neurons use to protect themselves from damage. 69 A further array of exotic compounds, whose description is well beyond the scope of this book (cytokines, interleukins, tumor necrosis factor, and so on), come into play as internal mood and beliefs confront external disease vectors and social tensions. An increased ratio of the right/left frontal lobe activation mentioned above as being associated with withdrawal and depression also correlates with immune suppression. 70

Interactions among the nervous, endocrine, and immune systems are not just of a negative sort. There is clear evidence that people who have a positive outlook and are confident of overcoming stressful barriers have higher blood levels of the T cells and natural killer cells that correlate with robust immune function. Some quite remarkable cases of mood- or belief-caused remission of the symptoms of illnesses such as cancer, arthritis, asthma, and acute depression have been documented. In many of these cases, a sham treatment called a placebo (Latin for "I shall please") is administered; it is believed by the patient to be an effective drug or therapy. A brightly colored pill containing sugar, a doctor in a white coat with a stethoscope, a shaman wearing a feather head dress---all can apparently induce a patient's immune system to respond to a disease crisis. Several experiments have paired an actual therapy, such as an asthma medicine, with an arbitrary pleasant flavor such as vanilla. Subsequent administration of the flavor alone can cause a significant increase in lung function. All of these examples indicate that expectancy---what the brain is telling the body it expects to happen---has a powerful effect on the actual outcome. The mind-body connection reaches down to the intricate details of our immune system biochemistry.

The endocrinologist Robert Sapolsky has conducted some fascinating studies on social roles, stress, and corticosteroid levels in baboon tribes in the Serengeti. 71 Like humans, these animals have to spend only a few hours a day gathering food and devote the rest of their time to driving each other crazy with elaborate social competitions. Measurements of physiological and hormonal changes relevant to stress show that virtually all of them are functions of rank. But it turns out on closer examination that rank is a less relevant variable than the baboon's "personality style." Those individuals who are best at telling the difference between threatening and neutral interactions, and who know clearly whether they won or lost, have the lowest glucocorticoid levels. It also helps, if you lose, to have somebody to take out your frustrations on. A second major factor in lowering corticoid levels is developing friendships. Does this sound familiar? Studies on humans yield the same results. And, as in humans, the animals who get agitated even when they see a rival at a safe distance are the ones that show hormonal correlates of stress.



Try another simple exercise on yourself. Close your eyes, breathe, and relax for just a few moments. Now, imagine that you suddenly see a car about to hit you. You are terrified and want to get away fast! Freeze your body right there. What muscles are tense? Your neck? Your back? These are the muscles involved in the startle response. This response takes about half a second, starting in the upper trapezius and sternocleidomastoid muscles and passing down the body. The curious thing is that the startle response is an extreme version of the slightly hunched-down "normal" postures that most of us display. Consider the possibility that the cacophony of input that we humans must put up with in this culture leaves many of us partially frozen in a startle response, or low-level anxiety. The sympathetic nervous system would be chronically activated, rather than being reserved for emergencies. 72 Back extensor muscles would be chronically overcontracted and front flexor muscles underused.


Our physiology clearly depends in part on our social context and can change with it. Social threats to our character have as a correlate threats to our integrated physiology. The effect of the social environment on stress extrapolates to susceptibility to, and recovery from, disease. 73 An extensive literature documents correlations between resistance to disease and social class, stress, social support, and the belief system of the individual. Stress and isolation correlate with reduced life spans. 74 Studies on communities of macaque monkeys and surveys on human communities show that close relationships buffer the immune system. Support systems that allow individuals to take life's ups and downs in stride enhance immune function and longevity. Depression, loss of status, or a sense of helplessness in an individual correlates with suppression of immune function. Our feelings of control versus helplessness exert a "top-down" control on our intimate biochemistry and neurophysiology. Comparison of subjects facing the same stressor shows that those who feel they have some control over the situation have lower cortisol levels than those who feel helpless. In short, there is a feedback loop between our social lives and our physiology. Our biological function is in a symbiotic relationship with our culture and society.


This chapter has chronicled several different aspects of the ill-defined array of experiences that we refer to as our emotions. Their roots lie much deeper in our evolutionary past than such recent inventions as language, and their link to our autonomic physiology---heart rate, skin gland regulation, and the like---distinguishes them from our "higher" functions of abstract and analytic thought. These higher functions, however, can be decisively biased by chemical modulators of emotion, originating in lower regions of the brain, that are released diffusely across the cortex. It appears that the vast array of values, hunches, inclinations, and rules that guide our selection of current actions from an array of options is programmed during our development by an interaction between higher and lower centers, as the more primitive circuits are recruited to assign different feelings or values to contexts that have been repeatedly encountered. These assessments then can work more rapidly than our self conscious cognitive processes to size up people or contexts that we encounter, as occurs when instantly (and, as it would seem, irrationally) we like or dislike someone we meet. They also can go to the extreme of becoming pathology, as in post-traumatic stress syndrome and obsessive-compulsive disorder. In both normal and abnormal circumstances, introspection techniques that enhance awareness of the distinction between emotional and self-conscious cognitive pathways in the brain can sometimes provide a useful tool for understanding or modifying one's behaviors. A maladaptive intrusion of our ancient emotional repertoire, causing damage that can include the immune system and brain, occurs if the stresses of daily modern life are permitted to activate chronically a nervous and endocrine stress system that evolved to deal with short-term life-and-death situations.

Our emotional minds, with their useful and their less useful elements, form another component of the mosaic of our modern minds, along with those outlined in previous chapters on plasticity, perception, and action. Now we are ready to tackle an issue distinctive to our human species: How is it that we are able to read and speak the words flowing past on this page? How do we acquire the faculty of language, the most important feature distinguishing us from the rest of the animal world?

Questions for Thought

1. Making a basic category of "the emotions" or coining terms such as "affective neuroscience" is a common-sense way of organizing descriptions of our experience, but it could be argued that these categories don't really correspond to fundamental entities---that emotions essentially are cognitions. Like a wide array of our sensory and motor behaviors, they reflect an analysis of some input to the organism that results in a behavior relevant to survival. Do you think emotions deserve description as a separate category? If so, why?

2. We could say that the main locus of linguistic cognition is our cerebral neocortex, the most recently evolved portion of our brain. How does this contrast with the location of the operations association with the emotions?

3. The psychologist Steven Pinker suggests that our lack of voluntary control over our emotions, and their link to basic physiological processes, are an evolutionary adaptations with a social purpose: to enhance the credibility of human social interchanges and make deception more difficult. Could such an idea be proved? Could there be other explanations?

4. Given the fact that many of our rapid and almost reflexive emotional behaviors can sometimes sabotage our effective performance in private and social contexts, do you think we would all be better off without these evolutionary vestiges of a former age---more like Dr. Spock of the television "Star Trek" series. If not, why not?

Suggestions for further general reading

Goleman, D. 1995. Emotional Intelligence. New York: Bantam. This best-selling book describes the mechanism and consequences of various kinds of "emotional hijacking." It also discusses the varieties and definitions of emotions.

Damasio, A.R. 1994. Descartes' Error---Emotion, Reason, and the Human Brain. New York: Putnam's. A stimulating discussion of how rational mind is built on top of emotional mind. This chapter draws heavily on Damasio's ideas about the origins of rational mind in emotional mind, adopts his distinction of primary and secondary emotions, and makes use of his concept of "somatic markers."

LeDoux, J. 1996. The Emotional Brain. New York: Simon & Schuster. In this book LeDoux, whose work is the basis of Figure 10-4, focuses on the neural mechanisms of fear.

Pinker, S. 1997. How the Mind Works. New York: Norton. In Chapter 6 of this book, Pinker discusses emotions from the perspective of evolutionary psychology and speculates that emotions are guarantors of the authenticity of our actions.

Sapolsky, R.M. 1994. Why Zebras Don't Get Ulcers: A Guide to Stress, Stress-Related Diseases, and Coping. New York:W.H. Freeman. This engaging book gives a fascinating summary of how body systems designed for episodically dealing with life-threatening stress can be chronically activated, with debilitating results.

Reading on more advanced or specialized topics

Panksepp, J. 1998. Affective Neuroscience. The Foundations of Human and Animal Emotions. New York: Oxford University Press. This is a textbook on the psychology and physiology of emotions.

Bear, M., Connors, B., & Paradiso, M. 1996. Neuroscience: Exploring the Brain. Baltimore, MD: Williams & Wilkins. Chapters 15 and 16 of this book describe the brain areas involved in modulating emotional behavior.

Davidson, R.J., & Sutton, S.K. 1995. Affective neuroscience: the emergence of a discipline. Current Opinion in Neurobiology 5:217--224. This is a more technical but very accessible review of brain mechanisms regulating emotions.

Pitkänen, A., Savander, V., & LeDoux, J.E. 1997. Organization of intra-amygdaloid circuitries in the rat: an emerging framework for understanding functions of the amygdala. Trends in Neurosciences 20:517--523.

Ekman, P., Davidson, R.J. 1993. Voluntary smiling changes regional brain activity. Psycholocial Science 4:342--345. An interesting experiment showing that brain activities for "fake" and "real" smiles are different, just as the facial muscle contractions are.

1. I realize that Descartes was talking epistemology, not biology, when he said that, and that, as the first to explore the physiology of the emotions, he is the intellectual ancestor of the science described in this chapter, but the pun was irresistible.

2. Davidson and Sutton, 1995.

3. Freeman, 1997.

4. Material in this paragraph is taken from Goleman, 1995d, Appendix A. This book, titled "Emotional Intelligence," provides a fascinating and lucid description of our emotional lives.

5. Cloninger, 1994; Eley and Plomin, 1997.

6. Ekman and Davidson, 1994, pp. 411-430.

7. Brief descriptions of these circuits can be found in Kandel et al, 1991, Ch. 47; or Bear et al. 1996, Ch. 16. A comprehension treatment of the brain and emotion is given by E.T.Rolls, 1999.

8. Ketter et al., 1996, give a review of limbic and prefrontal circuits involved in emotions and affective disorders.

9. Humphrey, 1992, pg. 162

10. Porges, 1995.

11. Seligman, 1994, Ch. 6.

12. Damasio, 1994, Ch. 7.

13. Darwin, 1872.

14. Goldsmith, 1991, Ch. 7.

15. Tooby and Cosmides, 1990

16. See Bear et al., 1996, Ch. 15, for a description of these diffuse modulatory systems.

17. Hen, 1996; Tecott and Barondes, 1996.

18. Tanda et al., 1997.

19. Self et al., 1996.

20. Angier, 1996.

21. Mirenowicz and Schultz, 1996.

22. Robbins and Everitt, 1966, provide a review of neurobehavioral mechanisms of reward and motivation. Neural substrates of prediction and reward are discussed by Schultz et al., 1997, and Schultz, 1997.

23. Giros et al., 1996.

24. The use of mutant mice in behavioral studies has become more problematic. Three separate laboratories using exactly the same inbred mutant mouse strains and attempting to use exactly the same breeding conditions and behavioral tests, found markedly different behaviors for the strains (Crabbe et al. 1999). This suggests that experiments characterizing mutants may yield results that are idiosyncratic to a particular laboratory.

25. See Robbins and Everitt, 1999, for brief review of how drugs of abuse might influence brain learning systems.

26. Panksepp, 1981. See Carter et al, 1999, for series of articles on the integrative neurobiology of affiliation.

27. Uvnas-Moberg, 1997.

28. see Carlson, 1991, Ch. 11, for the role of androgens in maternal and aggressive behavior; also Becker et al,1992, Ch. 9 for hormones and aggressive behavior

29. Reynolds, 1980, Ch. 10.

30. Devrets et al., 1977.

31. Damasio, 1994. The discussion in the following paragraphs is drawn from this book.

32. Bechara et al., 1996, 1997. Imaging experiments by Berns et al., 1997, have now shown that dorsal prefrontal and other brain regions can learn complicated sequences and respond to novelty in the absence of awareness. Thut et al, 1997, find that when a simple task is reinforced by monetary reward rather than just a simple "OK" that dorsolateral and orbital frontal cortex, along with regions of midbrain and thalamus, become more active.

33. Kawashima et al.,1993

34. Davidson and Tomarken, 1989; Davidson, 1992, Davidson and Sutton, 1995. See Davidson and Irwin, 1999, and Davidson et al, 1999, for more recent reviews on the functional neuroanatomy of emotion and affective style.

35. Goleman, 1996b.

36. This example is taken from Ornstein, 1991, pp. 82-83.

37. Kolb and Whishaw, 1990, Ch. 23; Zaidel et al., 1995

38. Hauser,1993

39. Goleman, 1995, Ch. 2, gives a lucid account of the anatomy of an emotional hijacking.

40. Damasio, 1994.

41. Ornstein, 1991, pg. 92

42. This work is reviewed by Goleman, 1995a.

43. Phelps and Anderson, 1997.

44. Bechara et al., 1995.

45. Adolphs et al., 1994; Clark, 1995; Maren and Fanselow, 1996.

46. Ohman, 1994.

47. Irwin et al, 1996.

48. Cahill et al., 1994.

49. See Kandel et al.,1991, Ch. 47, Fig. 47-2 for more discussion

50. Goleman, 1995, Appendix C. is an excellent description of the neural circuitry of fear. See also Killcross et al, 1997, for description of how different types of fear-conditioned behaviors are mediated by separate nuclie within the amygdala.

51. Drevets et al, 1992.

52. Relevant references are given in Davis, 1992 and Davidson and Sutton, 1995.

53. Kalin, 1993

54. LeDoux, 1994. The drawing is a modification of one in this article.

55. Waldholz, 1993

56. Tovee, 1995

57. Niedenthal and Kitayama, ed. 1994

58. Ekman, 1992. See also Izard, C.E. (1971).

59. Ekman, 1973

60. Ekman, 1992

61. Ekman and Davidson, 1993.

62. Sapolsky's very engaging book, "Why Zebras Don't Get Ulcers", 1994, gives a detailed account of this issue. Ornstein and Swencionis, 1990, is a useful source for other points made in this section.

63. We can have another response to stress and challenges, parasympathetic rather than sympathetic over activation can lead to a `freezing' sort of reaction that slows the heart, shutting down the body rather than mobilizing it. Such a slowing down may underlie an phenomenon seen in some primitive cultures, were individuals have been observe to slow down and die after being cursed or ostracized by their group.

64. C.Pert, Ch.13 in Ornstein and Swencionis, 1989; Pennisi, 1997.

65. Hosoi et al., 1993.

66. McEwen and Sapolsky, 1995. Sapolsky, 1996.

67. Sheline et al., 1999.

68. Marx, 1995.

69. reviewed by Fischman, 1993

70. Davidson and Sutton, 1995.

71. Sapolsky, 1994, Ch. 13.

72. see also Ornstein, 1991, pp. 90-95

73. Chapter 10 in Becker et al (1992) deals with the effect of stress on the immune response.

74. Points in this paragraph are covered by Goleman's, 1995, book.

Home   |   Mind Blog   |   Lectures & Writings   |   Music   |   About Me   |   Contact

©Deric Bownds • All Rights Reserved
Counter 1