Scarlet & Grey
Ohio State University
School of Music


Joseph LeDoux

Notes by Matthew Bester


Music 829
May 22, 2001

Joseph LeDoux. 1996. The Emotional Brain: The Mysterious Underpinnings of Emotional Life. New York: Simon & Schuster.

"The Emotional Brain provides an overview of my ideas about how emotions come from the brain. It is not meant as an all-encompassing survey of every aspect of how the brain produces emotions. It focuses on those issues that have interested me most, namely, issues about how the brain detects and responds to emotionally arousing stimuli, how emotional learning occurs and emotional memories are formed, and how our conscious emotional feelings emerge from unconscious processes." (p. 9)

LeDoux focuses on a strong scientific approach, at the expense of 'intuitive' beliefs. Emotions are viewed as biological functions of the nervous system, particularly the brain itself. In underscoring the importance of the brain in relation to various psychological theories of emotion, he notes, "there are many possible solutions to the puzzle of how emotions might work, but the only one we really care about is the one that evolution hit upon and put into the brain." (p. 13)

LeDoux came to emotion research from studying split-brain patients, whose two brain hemispheres were severed as a last resort against epileptic seizures. During the course of this research, LeDoux and his Ph.D. advisor discovered a patient who, when presented with stimuli to the right hemisphere, could not describe the stimulus (a left hemispheric function) but could describe the emotional impact that the stimulus had on him. This seemed to reveal a "fundamental psychological dichotomy - between thinking and feeling, between cognition and emotion." (p. 15)

Because technology was not sufficiently advanced to study the human brain thoroughly when LeDoux began his pursuit of emotions in the brain, LeDoux has largely relied on the study of animals, particularly rats, for his research. He defends this practice from a strong evolutionary perspective.

LeDoux sets out his primary themes in the first chapter as follows:  The proper level of analysis is that at which a psychological function is represented in the brain. "The word 'emotion' does not refer to something that the mind or brain really has or does." (p. 16) There is no 'emotion' faculty in the brain, nor is there a specific location where emotion is processed. There must be a specific system for each emotion.  Brain systems that generate emotional behaviors are highly conserved through many levels of evolutionary history. Animals with backbones and brains share similar emotional behavioral systems as well as biological imperatives of procreation, food, shelter, etc.  The occurrence of these systems in a being that has the capacity for conscious awareness leads to conscious emotion feelings. This capacity definitely exits in humans, but it is questionable whether it exists in other animals. "If we do not need conscious feelings to explain what we would call emotional behavior in some animals, then we do not need them to explain the same behavior in humans." (p. 17)  Conscious feelings by which we know emotions are 'red herrings' in the scientific study of emotions. The conscious state of fear, for instance, is not as important as the system that detects and responds to danger in the first place.  If emotional feelings and responses are in effect from a 'common underlying system', studying the 'objectively measurable emotional responses' can tell us about this mechanism. This leads again to the importance of animal studies for understanding human emotion.  Conscious feelings of emotion are no different from other states of consciousness. "What differs… is not the system that represents the conscious content… but the systems that provide the inputs to the system of awareness." (p. 19)  Emotions happen to us. They are not things that we will to occur. Faking an emotion is futile.  Emotions, once they have occurred, "become powerful motivators of future behaviors." (p. 19)

The Importance of the Unconscious to Cognition and Emotion

One of LeDoux's major contentions is that both cognition and emotion seem to operate unconsciously. As he points out, "Although our emotions are at the core of who we are, they also seem to have their own agenda, one often carried out without our willful participation." (p. 22)

However, emotion and reason have been separated since as far back as the Greek philosophers. Even the relatively new field of cognitive science largely has ignored emotion in favor of thinking, reasoning, and intellect. Although the field of artificial intelligence (AI) increasingly has acknowledged the importance of the unconscious to cognition, it has largely been an emphasis on unconscious processes rather than unconscious content. Functionalism (the belief that achieving the same cognitive result regardless of the 'hardware' employed implies the existence of the same underlying process) leaves out the importance of consciousness (hence emotions).

However, studies have shown that rationality in humans often is not achieved by following formal laws of logic. Johnson-Laird, for instance, claims that we use 'mental models', hypothetical models drawn from past experiences or imagined situations. Others assert the irrationality of many decisions. "If cognition is not just logic, and is sometimes illogical, then emotion might not be as far afield from cognition as it was initially thought." (p. 36) Indeed, many emotions seem to be the product of a sort of 'evolutionary wisdom' (as an example, consider animals who recognize a predator as a threat immediately upon first encounter). Antonio Damasio, whom we encountered earlier, also cites the rationality of emotion.

Another reason that emotions were not included in the cognitive revolution has to do with the fact that they were seen as subjective states of consciousness. Conversely, LeDoux believes that emotions are best seen as the "end result of information processing occurring unconsciously." (p. 37) LeDoux is led to the conclusion that both emotion and cognition can be studied using similar concepts and experimental tools, as they "both involve unconscious information processing and the generation of conscious content on the basis of this processing." (p. 38)

LeDoux attacks the functionalist assertion that the hardware of the information processing system is unimportant on the basis that the body and its responses are of obvious importance to the experience of emotion. A conscious computer could never experience emotion as we know it, for it lacks the appropriate composition, conditioned by eons of biological evolution.

Comparing the Prevailing Psychological Theories of Emotion

LeDoux sets out to reconcile his understanding of the brain and the importance of unconscious thought with the various psychological theories of emotion. He believes that appraisal theories of emotion come the closest to getting it right. The evaluation of a stimulus is the first step. However, because these appraisals occur unconsciously, LeDoux points out two failings of traditional appraisal theories:  Most appraisal studies have relied heavily on self-reports.  At the same time, appraisal theories have often overemphasized conscious cognitive processes in emotion.

LeDoux sees the work of Robert Zajonc (1980) as a turning point. Zajonc demonstrated that emotion has primacy over and is independent of cognition (experiments showed that participants showed a preference for previously encountered stimuli even when they did not recall having encountered the stimuli before). It appears that emotions are more easily influenced when we are unaware of the manipulation. For instance, studies conducted with subconscious emotional priming prior to a neutral stimulus (e.g., a Chinese ideogram) resulted in emotional manipulation.

LeDoux therefore believes in adopting the term 'mind science' rather than cognitive science. In such a discipline, cognition and emotion are given equal billing (but remain separate). LeDoux's reasons for studying cognition and emotion as separate but interacting mental functions and brain systems are as follows:  Brain damage sometimes causes patients to lose the capacity to assess the emotional significance of some objects without losing the capacity for perception of the object.  The emotional meaning of a stimulus can begin to be appraised before perceptual systems have fully processed the stimulus.  Memories of emotional significance are registered, stored, and retrieved differently from cognitive memories.  Systems that perform emotional appraisals are connected to systems involved in emotional responses. Cognitive systems, on the other hand, are not as tightly coupled, thereby increasing their flexibility.  This linkage means that emotional appraisal is often accompanied by direct physiological responses, which become part of a conscious awareness of the emotion. (This is a potentially powerful reconciliation of the Jamesian perspective with appraisal theories.)

Because of the importance of the unconscious, LeDoux is wary of psychological studies of emotion that rely too heavily on verbalizations of emotion. "Given that so much work on unconscious processing (cognitive and emotional) has focused on verbal processes, we probably have a highly inaccurate picture of the level of sophistication of unconscious processes in humans." (p. 71) He points out that humans were non-verbal long before we were verbal. As a result, we can hope to learn much about the subconscious from studying animals.

The Limbic System: Is it Real?

For much of the previous century, the Holy Grail of brain research has been to locate the area in the brain responsible for emotion. Since the mid-twentieth century, the limbic system has been assumed to be the culprit. However, LeDoux's studies have led him to conclude that the limbic system theory is wrong. Indeed, the limbic system may not even exist. "Brain regions, in short, have functions because of the systems of which they are part. And functions are properties of integrated systems rather than of isolated brain areas." (p. 78)

Historically, Paul MacLean cemented the importance of the limbic system in 1949. It was then known as the smell brain, or rhinencephalic areas, which MacLean renamed the visceral brain and later the limbic system. Particularly important in this theory was the hippocampus, which was dubbed the 'emotional keyboard.' To MacLean, the primitiveness of the hippocampus area in contrast to the more recently evolved neocortex accounted for our difficulty in understanding and verbalizing our own emotions. MacLean also espoused the 'triune brain' concept, whereby the reptialian, paleomammalian, and neomammalian brains are all present in humans and primates, but only the first one or two are present in lower animals. This theory implied the importance of evolutionary development for understanding emotion in various animals.

However, anatomists showed in the 1970s that many 'lower' mammals have brain areas that meet the criteria of the neocortex. It is no longer possible to say that some brain areas are older than others by reference to 'older animals' (evolutionarily speaking).

Other recent brain studies have further damaged the limbic system theory. Advances in neuroscience have shown that the hypothalamus is connected to all parts of the brain, so localization of emotion processing in the limbic region (traditionally defined) is not viable. At the same time, patients with brain damage to the hippocampus demonstrate no consistent emotional impairment but do have problems with conscious or declarative memory. As a 'primitive' brain, the limbic system was not supposed to take part in such functions. LeDoux concludes that there probably is not one emotional system in the brain but many and that emotions therefore must be studied one at a time. LeDoux primarily has focused on fear.

Universal Emotions and 'Learned Triggers'

LeDoux believes that "some emotional systems in the brain are essentially the same in many of the backboned creatures, including mammals, reptiles, and birds, and possibly amphibians and fishes as well." (p. 107) This cross-species universality leads him to reject the strong claim of social constructivism. In LeDoux's assessment, the fact that some emotions have no clear translatable equivalent in other cultures is not a sufficient rejection of the basic emotions (and combinations thereof) proposed and studied by Ekman, Izard, Tomkins, Frijda, and Plutchik. He holds up Ekman's concept of display rules in conjunction with universal emotion expressions as going a long way towards explaining individual and cultural differences in emotional behaviors.

LeDoux suggests that "the most practical working hypothesis is that different classes of emotional behavior represent different kinds of functions that take care of different kinds of problems for the animal and have different brain systems devoted to them." (p. 127) Each system, then, must have its own 'natural triggers' for the unconscious appraisal process (e.g., the example of an animal recognizing a predator as dangerous on first encounter). However, there also are learned triggers. As a result, nature and nurture are equally important to emotional development.

Focusing on Fear

LeDoux defines the fear system not in terms of the system that results in the experience of fear but purely as a defensive mechanism that produces responses to maximize the possibility of survival. In order to study the flow of information in brain neurons in the presence of fear-inducing stimuli, LeDoux and others use 'fear conditioning' as a behavioral tool, whereby meaningless stimuli are turned into warning signs by past experience with similar situations (a variation of Pavlov's famous conditioning experiments). LeDoux points out that the freezing response to an unconditioned or conditioned stimulus is the same and must therefore by hard-wired. The physiological response of freezing is not learned but a product of evolutionary emotional development.

A common set of brain structures and pathways have proven to be important for fear across many species regardless of the stimulus. Charting these structures and pathways is often done by lesioning specific areas of the brain and assessing the animal's behavioral changes in response to stimuli. In studies of rats, it was found that damaging the auditory cortex caused no change in fear response, while damage to the thalamus did prevent fear conditioning. Thus conscious cognition was not necessary for conditioning. Later tests showed that disconnection of the amygdala similarly impaired conditioning. As a result, it has been concluded that the lateral nucleus of the amygdala, rather than the auditory cortex, is the next destination after the thalamus for conditioning stimuli, thereby bypassing cognitive functions.

Later studies showed that the central nucleus of the amygdala had connections with brain stem areas that control heart rate and other autonomic nervous system responses. Subsequent research in which the central nucleus of the amygdala was impaired resulted in interference with every measure of conditioned fear. The lateral nucleus of the amygdala therefore appears to be the input station, while the central nucleus is the output station.

Despite the bypassing of the auditory cortex for simple fear conditioning, it has become clear that the cortex does become involved for more complex conditioning (e.g., compare the processing of a single tone as a stimulus versus an entire sentence). Additionally, the hippocampus, with its role in long-term memory, has been shown to be important for recognizing stimuli that are contextual to a conditioned stimulus (e.g., rats whose conditioned response has been cancelled in a cage other than that in which the conditioning took place will have a recurrence of the response when returned to the original cage, even without the presence of the original stimulus). Significantly, the amygdala is a hub for all three types of stimuli. It appears to be the appraisal center for fear.

Studies in humans with epilepsy as well as those with brain damage to the amygdala suggest that the amygdala plays a significant role in fear responses for humans as well as other mammals. In one remarkable case, a patient with amygdala-specific damage was able to detect emotional expressions on faces except when they showed fear.

As was mentioned above, the connection between the thalamus and the amygdala is the most direct and therefore the fastest. It is not surprising, then, that our brains err on the conservative side when encountered with fear stimuli, no matter how much we may be consciously aware of the situation (think of Dr. Huron's door-slamming example). One byproduct of this conservatism is that reaction buys time for cognition to shift from reaction to action. The prefrontal cortex seems to be essential for this transition. Emotional plans, as LeDoux calls them, allow us to be emotional actors rather than just reactors. 'Coping' can be seen as the cognitive planning of voluntary actions in the midst of an involuntarily elicited emotion reaction.

Implicit vs. Explicit Memory

Crucial to the mechanisms of conditioned responses and emotional learning is the concept of implicit memory. It now appears that different brain systems are responsible for declarative or explicit memory vs. implicit emotional memory involved in conditioning. Many amnesiacs, for instance, retain the ability to learn tasks and can still be primed without the conscious memory of previous exposures. Explicit memory relies primarily on the temporal lobe memory system, including the hippocampus, while there are multiple systems for different types of implicit memory (skill learning, conditioning, etc.).

It has been shown that both implicit and explicit stimuli can arouse the amygdala. Thus, there appear to be different systems for forming implicit emotional memories and explicit memories of emotions. Indeed, the amygdala seems to functionally mature before the hippocampus, which may account for our general lack of long-term memories from before the age of 3 at the same time that emotional traumas may affect us from before that age.

It also has been shown that adrenaline, often released in emotion-inducing events, helps to sharpen the creation of long-term memory. This suggests that emotional memories are stronger than non-emotional ones. However, such explicit memories will not always be accurate or complete. For one thing, the formation of a long-term memory depends on what was being attended to at the time of perception (e.g., a victim of an armed robbery may focus primarily on the gun being waved at his or her face rather than on details of his or her attacker or surroundings). At the same time, the recall of these memories is colored by intervening events as well as by the state of the brain at the time of recall.

Despite differences in circuits for different types of memory, however, there is reason to believe that memory and learning are governed by a common chemical process affecting the relationships of neurons on a molecular level. This basis can account for the emotional quality of explicit memories of past emotions.

Anxiety Disorders: Special Cases of the Fear Response

LeDoux believes that anxiety disorders are a result of the fear system breaking loose of the cortical controls that keep primitive impulses in check. Anxiety typically is distinguished from fear by the absence of an external stimulus that elicits the reaction. When anxiety or fear exceeds or is more persistent than what is reasonable under the circumstances and impede normal life, a disorder is present. Recent research has shown that fear conditioning contributes significantly to such disorders.

In order to understand why some conditioned responses are easily cancelled while other responses are not, Martin Seligman has advanced the concept of evolutionary preparedness. This may explain the hard-wired natural triggers associated with many animals. Furthermore, preparedness suggests that amygdalas may be pre-wired with certain connections that facilitate survival responses and signals. Facial expressions in humans and other primates may be such a hard-wired signal. Unfortunately, it is believed that some humans may be super-prepared for various stimuli, thereby increasing the likelihood of the development of phobias.

Recent research has shown that stressful events can cause malfunctions in the hippocampus, which may account for some patients' lack of explicit memories of traumatic events. Additionally, prolonged exposure to stress has been shown to shrink the hippocampus and impair memory. This runs counter to the Freudian concept of repression in traumatic circumstances; rather, the impairment of the hippocampus prevents long-term explicit memories from forming altogether in these situations. However, stress does not seem to affect the amygdala; in fact, amygdala processes may be enhanced under intense stress, including conditioned fear. As such, powerful implicit memories may be formed while explicit memories are impaired. This suggests that "stress shifts us into a mode of operation in which we react to danger rather than think about it." (p. 247)

Other studies have shown that unconscious fear memories established through the amygdala remain present for life. Conditioned responses can become extinct by mitigation by the medial cortex, but stress or other contextual memories can cause the conditioned response to return without warning. Seen in this light, therapy can be seen as a strengthening of control over the amygdala by the cortex.

Consciousness: The Final Frontier

In the last chapter, LeDoux turns to the problem of consciousness and feeling, or what he describes as subjective emotional experience. For LeDoux, two things are necessary for such a conscious experience: a specific emotion system and a conscious awareness of its activity.

LeDoux asserts that this awareness takes place in the short-term 'working' memory system, a cognitive workspace that makes reasoning possible. The working memory system is dependent on the lateral prefrontal cortex, which only exists in primates (and is considerably larger in humans). Many contemporary cognitive scientists have described consciousness itself as the awareness of what is in working memory. As LeDoux explains, "feelings come about when the activity of specialized emotion systems get represented in the system that gives rise to consciousness, and I'm using working memory as a fairly widely accepted version of how the latter might come about." (p. 283)

Significantly, by way of connections with specialized short-term buffers, long-term memory, and other networks of the frontal lobe, the amygdala has the ability to influence the information content of working memory. In addition, emotional reactions typically are accompanied by high levels of cortical arousal, which makes information processing systems hypersensitive to stimuli. The amygdala itself has been shown to be an influence on the release of a neurotransmitter that heightens cortical arousal.

Bodily feedback also is important to the conscious awareness of emotion. Although visceral responses in the brain are too slow to be the primary determining factor of emotion (it takes 1-2 seconds for signals to travel from the brain to the viscera and back again), other somatic responses are quicker and can have an impact on the conscious awareness of emotion. Taken together, possibilities exist for emotion-specific feedback.

LeDoux concludes, saying, "Emotions evolved not as conscious feelings, linguistically differentiated or otherwise, but as brain states and bodily responses. The brain states and bodily responses are the fundamental facts of an emotion, and the conscious feelings are the frills that have added icing to the emotional cake." (p. 302)

Conclusion and Assessment

LeDoux offers a compelling model for emotional response as a primarily unconscious nervous system activity. However, this approach may be tipped too far in the direction of the subconscious when applied to humans, for whom conscious cognition is a continuous reality. As LeDoux himself points out, there are connections from the cortex to the amygdala. It therefore is not unreasonable to expect that conscious cognitive stimuli may have more of an effect on emotional response and experience in humans than in other related animals.

Nevertheless, the biological basis of this model provides valuable links between the four major perspectives of emotion that have been discussed thus far. In particular, it provides the starting point for a reconciliation of the Darwinian, Jamesian, and Cognitive perspectives, made possible by the existence of highly evolved pathways for unconscious appraisal and response as well as bodily feedback processes that can affect the conscious experience of emotion. In addition, the capacity for emotional learning and memory, independent but working in conjunction with explicit memory, helps to explain individual and cultural differences in emotion responses that are at the core of the social constructivist paradigm. A grand unification of the competing perspectives may yet be possible as more is learned about the brain and its relation to the body.



This document is available at http://csml.som.ohio-state.edu/Music829D/Notes/LeDoux.html