Chapter Overview

The importance of emotion for human functioning is obvious. On an immediate level, love or fear may absorb our attention and dominate our conscious experience, leaving us unable to deal with other matters. This may have enormous survival value; faced with an angry bear, it is not very adaptive to contemplate one's stock portfolio. From a longer-term perspective, emotions are behavioral and physiological specializations that have evolved because they increase the chances that a particular species will survive. The process of natural selection that drives evolution means that organisms with certain characteristics will survive and transmit these characteristics to their offspring, whereas organisms with less-adaptive characteristics will fail to do so. The net result of evolution's tinkering with the gene pool is a combination of change and conservation of characteristics, depending on the demands of the environment. Although emotional systems have evolved, their evolution has been strikingly conservative. This means that emotional processes are more similar across different species than are most other psychological processes. This conservatism is understandable when one considers that maladaptive characteristics in a system so closely keyed to survival and procreation may rapidly lead to the death of the individual with those characteristics. When it comes to survival mechanisms, maladaptive characteristics are ruthlessly expunged from the gene pool.

For all of its importance, emotion and its neural mechanisms have been less studied than most other areas of psychological functioning. This is partly due to the difficulties inherent in studying emotion. In particular, the subjective nature of conscious emotional experience has alienated many researchers. This is partly a by-product of behaviorism and its repudiation of subjective experience as a fruitful domain of study. But the study of emotion in its own right has also been largely excluded from the cognitive psychology that has dominated psychology for the past three decades. Instead, emotion has been approached as if it were a province of cognition.

An example of cognitive approaches to emotion is Schachter and Singer's (1962) cognitive arousal theory of emotion. In their classic study, Schacter and Singer injected subjects with epinephrine and then exposed them to a social situation that was likely to evoke either positive or negative emotions. In this case, the social situation was the presence of a confederate of the experimenter who was either in a buoyant, happy mood or an angry, hostile mood. The critical manipulation of the experiment was that some of the subjects were told that they had received epinephrine and that they should expect to experience autonomic arousal (e.g., increased heart rate), whereas others were not told that they were receiving epinephrine and that they were about to experience sympathetic arousal. Most subjects experienced the same quality of emotion (positive or negative) as that displayed by the confederate. The most important finding, however, was that subjects who were not told that they were receiving epinephrine experienced stronger emotions than those who were told to expect sympathetic arousal. Uninformed subjects apparently attributed their arousal to their emotional state, and this attribution enhanced the intensity of their emotional experience. Informed subjects, in contrast, attributing their arousal to the epinephrine and not to their emotional state, experienced less-intense feelings. The key factor, the investigators concluded, was the subjects' cognitions about their arousal, rather than the arousal per se. In fact, it was subsequently shown that a subject's belief that he or she is being aroused influences emotional experience, even when that belief is based on false information (Valins, 1966).

These were seminal studies and are representative of a large number of studies emphasizing the importance of cognitive interpretation as a factor contributing to emotional experience (Figure 11.1). The cognitive approach to emotion has been useful in that cognitive attributions and interpretations clearly do influence what we feel (Arnold, 1960; Lazarus, 1966). In addition, applications of this approach in the form of cognitive therapy have been effective in the treatment of mood disorders (Beck, 1967; Kelly, 1955). Nevertheless, cognitive approaches to emotion leave out aspects of emotion that lie outside the cognitive domain, aspects that made their appearance in the course of evolution long before cognition. These essentially emotional systems are conserved in the biologically based survival mechanisms of present-day animals, and a biological approach to emotion therefore offers the possibility of attaining a more complete and balanced understanding than is attainable through the exclusive use of the cognitive perspective.

A biologically based approach also affords the possibility of integrating cognitive and emotional factors into a more unified view of human decision making. Emotion and cognition ultimately work together to mediate higher-order human behavior. This idea, espoused by poets and philosophers for centuries, has recently gained empirical support from investigations in the neurobiology of emotion. There is evidence, for example, that the impairment in reasoning seen in patients with orbital frontal lesions is not due to a primary deficit in thinking but, rather, to the unavailability of emotional input to the reasoning process. Such patients have been shown to reason well in laboratory tests that do not require an emotional contribution, only to be very impaired in real-life decision-making situations that depend on being in touch with one's emotions for their adaptive solution (Damasio, 1994). This interdependence of emotion and cognition has led some to suggest that, in studying the neural basis of psychological functioning, the much-used term cognitive neuroscience is too restrictive and biased, implying some kind of priority to cognition over other processes, including emotion. The term mental neuroscience has been proposed as an alternative (LeDoux, 1996). The term psychic neuroscience might be preferable because it would seem to be burdened with even fewer assumptions about the basic elements of psychological processes.

Whatever we call it, making the brain the focus of the investigation of psychological processes has the advantage of providing a perspective that may help free us from some of the biases inherent in cognitively based perspectives. In this chapter, we will address the question of what the brain tells us about emotion. This approach will allow us to avoid some of the assumptions inherent in cognitive approaches to emotion. Framing the question in this way also provides an opportunity to apply what is sometimes called reverse engineering (Pinker, 1995). This is the idea that, because evolution has "engineered" (through the process of natural selection) a highly adaptive biological mechanism for a certain psychological process, a strategy making that mechanism the focus of investigation is likely to reveal much about the nature of both the brain mechanism and the psychological process it mediates.

Emotion can be divided into three components: body changes mediated by the autonomic nervous system and the endocrine system (such as increased heart rate), behavior, and subjective experience. The last of these domains presents a problem. Whereas body changes and behavior can be objectively measured, feelings, being subjective, cannot. As mentioned earlier, this has had a dampening effect on the experimental study of emotion. The position taken here (and advocated by a number of neuroscientists studying emotion, e.g., LeDoux, 1996) is that, as important and interesting as feelings are, it is not necessary that they be tackled directly in order to study emotion. Analogously, cognitive psychology has found ways to study a wide variety of processes, from mental rotation to priming, through the analysis of observable data. According to this approach, the study of emotion can focus on understanding how various specialized brain systems process information and produce responses that are characteristic of various classes of emotion. This can be done without reference to conscious emotional experience.

In the sections that follow, we will first discuss some of the early attempts to understand the relationship between brain and emotion. Some of these are enticingly inventive and have stimulated some illuminating experimentation. In the end, however, they leave us disappointed. One reason for this is that they attempt to deal with emotion as if it were a global phenomenon. We have seen that when Karl Lashley treated memory as a global phenomenon, he was unable to devise experiments that shed light on its neural basis. It was only as it was realized that memory comprises a number of different systems that it became possible to gain some meaningful understanding of the neural basis of each of these systems (see chapter 10). There is every reason to believe that the search for the neural mechanisms of emotion will follow a similar course, and, in the second section of this chapter, we examine the neural basis of conditioned fear as a model system. In the third section, we examine emotional memory and compare it with explicit memory for emotion. We will find that different systems underlie these two domains of memory. In the fourth section, we examine the role of the cerebral cortex in emotion, and, in the fifth section, we discuss the interactive relationship between cortex and amygdala in the higher-order mediation of emotion-based action. Finally, in the last section we will briefly tackle the problem of emotion and consciousness.