Chapter 5:     Mood

The chapter steps through the findings of neuroscience concerning mood. It concludes that, according to current findings, the left prefrontal cortex (PFC) generates and processes the mood of optimism; and the right PFC, the mood of pessimism.  Further, these findings support the conclusion that each of us typically exhibits a characteristic asymmetry dominance pattern in the PFC. In other words, some of us are usually left PFC dominant; others, right PFC dominant; still others, left and right PFC balanced.  But all of us are somewhere along the continuum from extreme left dominance to extreme right.

As we saw in Chapter 3, mood is also at core of the Enneagram theory of personality. So the current chapter lays the groundwork for linking that theory with the brain. But as Part One shows, that theory is far richer than simply a description of a few mood states. There is more to personality than mere mood.

So, in addition to analyzing the neuroscience findings on mood, this chapter also touches on consciousness and cognition. These additional subjects serve to flesh out the rich elements of personality.

At the same time, in dealing with these broader subjects together with mood, this chapter stumbles upon some provocative suggestions for how neuroscience might develop. Not that most, if any, neuroscientists would heed the suggestions here. But all the same, these suggestions may be interesting.

The introduction to Chapter 3 warned that the text to follow was going to be "dense". The same caution applies here. So check your mental reserves before forging ahead. With reserves filled, we are now ready to dive into the brain in search of mood.


Although this chapter deals with mood (optimism and pessimism) and the PFC, it would be misleading to suggest that processing mood is the dominant function of the PFC. Recent research is suggesting that only a small area (called the orbitofrontal cortext) of the PFC is responsible for mood. So the PFC dishes up much more than just mood. But what else?

Perhaps the most important function of the PFC is to house "working memory".[1]  In Synaptic Self, Mr. LeDoux devotes an entire chapter to working memory.[2]  "Working memory" is the name neuroscientists give to the common notion of "short term memory".  But "working" memory is a better name because it suggests more about the stored information than merely its temporary nature.  The name also highlights the "work" that is performed upon this information.  Mr. LeDoux calls this work the "executive functions" and says these functions are ultimately used for activities like planning, problem-solving, and behavioral control.[3]

Working memory is important because it is intimately related to consciousness. As Mr. LeDoux says: "The stuff we are conscious of is the stuff that working memory is working on."[4]  That is, to my knowledge, we still don't know where consciousness comes from nor why it shifts from one thought to another. But in the early 21st century, we do suspect that whatever consciousness is thinking about, that stuff is stored and processed in the working memory area of the PFC.

In this chapter, the notion of the "conscious field" will become important, particularly in how it relates to the rest of the brain. For understanding this notion, think about an old-style submerged submarine, including almost everything and almost everyone in it.  The conscious field is the captain who is operating the periscope.  The periscope is something Mr. LeDoux calls "passive awareness"[5]. And everything else about the sub outside of the captain and the periscope is "unconscious processing." The captain can move passive awareness (the periscope) to scan the environment. He can also order some of the unconscious processing (the crew) to perform work.  But there's much more happening in the sub about which the captain has no idea, and over which he has only tenuous and marginal control, or, more commonly, no control at all.

The above metaphor explains that what is going on in our conscious minds at any moment represents only a small part of what is going in the rest of the brain. As you will see in the remainder of Part Two, mood and fear are processes that seem to be part of the "rest of the brain", outside of consciousness. This doesn't mean we can't become conscious of our moods and our fear. Rather, it means that we're not necessarily conscious of them.

With this introduction, we are now ready to dive into PFC asymmetry, a milestone on the road toward optimism and pessimism.

Trait Asymmetry

As mentioned in Chapter 4, the PFC is divided into a left and right side, and, generally speaking, activation of the two sides is asymmetric. That is, some times the left PFC is more active, other times the right PFC is more active, and still other times activation is more or less symmetric.

This section looks at settled findings showing that individuals differ in a systematic way according to PFC asymmetry. So PFC activation for some people tends to skew left, for others it skews right, and for still others it tends to be more symmetric. These systematic differences between individuals is the topic of "trait asymmetry".

Mr. Davidson is a leader in the field of neuroscience with respect to PFC asymmetry.  Specifically, Mr. Davidson's expertise lies in the area of PFC asymmetry as it applies to emotion, rather than cognition.  Below, this chapter explains the difference.  But for now, the focus is on the simple fact of asymmetric dominance in activation.

Over the past ten years or so, Mr. Davidson has published "review" papers that summarize not only his own research, but also try to capture the state of the field.  As of this writing, his most recent review papers are a 2004 commentary entitled "What does the prefrontal cortex "do" in affect: perspectives on frontal EEG asymmetry research"[6] and a 2003 article entitled "Affective neuroscience and psychophysiology: Toward a synthesis".[7]  These papers offer a useful summary of Mr. Davidson's views circa 2004.

Mr. Davidson explains that PFC asymmetry is a trait-like measure.[8]  Being a trait-like measure means that this asymmetry is reliable and stable over time.[9]  For example, as you are reading this page, your PFC is likely exhibiting a certain pattern of asymmetry.  That means there is a certain balance between left and right activation.  According to the trait theory, that same balance will be present in your PFC later today when you brush your teeth, and next week when you watch television.  This notion is akin to handedness.  Perhaps you write with your right hand.  If you are using your right hand to write today, you will probably be using it tomorrow, the day after, and so on.

Now testing for handedness is easy.  All that needs to be done is to observe what hand the subject chooses for activities like writing.  PFC asymmetry is harder to measure.  It requires some equipment.

The most basic equipment for measuring PFC asymmetry is EEG.  Because the PFC is located just under the skull at the front of the head, EEG works well in detecting activity in this region.  Because trait measures are relatively stable across time, fMRI and PET are also useful for these measures.  But, at present, the use of EEG seems to predominate in this area likely because EEG is far less expensive.

With the use of EEG, individuals can be distinguished as "lefties", "righties", and "middle" folks in terms of PFC activity. In some experiments, Mr. Davidson has studied "extreme left [lefties] and extreme right [righties] frontally [PFC] activated subjects."[10]  But, of course, some individuals are only mildly left or right dominant.  The notion is that, for any individual, this asymmetry measure can fall anywhere along the continuum from extreme left to extreme right.

Talking about "extreme lefties" might make it sound like such individuals exhibit activity in their left PFC, but little or no activity in their right PFC. For normal people, this is not the case.  In fact, the truth is not even close to this.  Researchers explain that "the difference between hemispheres is relatively small compared to the overall magnitude of activity in each hemisphere."[11]  This slight difference in activity is on the order of one to five percent.[12]

Given that the difference in activity between left and right PFC for an extreme lefty and for an extreme righty is so slight, it may seem surprising that the emotional characteristics of such people are so different.  But this is exactly what the remainder of this chapter reveals.

Not only does trait asymmetry represent a relatively slight difference in PFC activity, but those slight differences are not rigidly stable.  The word "trait" indicates that a particular pattern of asymmetry persists over time.  But this doesn't mean that this pattern exists at every moment in time.

Researchers use the phrase "test-retest reliability" to refer to the practice of taking readings of the same people on different days and comparing the two sets of results.  In some of his experiments, Mr. Davidson has observed a test-retest reliability of 65% to 75%.[13]  Other researchers have recorded scores of 44% to 71%, and 66%.[14]

In some of his studies, Mr. Davidson has grouped individuals into one of three PFC asymmetry categories: extreme left, extreme right, and middle (i.e. relatively even left and right).  The test-retest data above suggests that some non-trivial percentage of individuals fell into different groups (left, right, or middle) on different days.

But I am unaware of studies looking at the individuals who failed to show test-retest consistency. [15] Such research might ask questions like: If an individual exhibited, say, left PFC dominance on one day of the experiment, what pattern of dominance did he exhibit on the other day?  Was it middle, or right, or merely a little less or more left?  Part Three suggests this avenue of research may prove promising.

State versus Trait Asymmetry

Trait asymmetry is one way of looking at PFC asymmetry.  Another way of looking at it is known as "state asymmetry". As researchers explain:

[Trait asymmetry] involves correlating resting EEG activity with trait-like phenomena such as temperament or psychopathology, or with state fluctuations in emotion.  [State asymmetry] involves correlating state fluctuations in frontal EEG asymmetry with changes in emotional or motivational state.[16]

A simpler way to characterize the difference between the two approaches is to recognize that trait asymmetry focuses on differences between individuals, whereas state asymmetry focuses on similarities between individuals. [17] 

In a typical trait asymmetry experiment, the first thing that happens is that resting "baseline" PFC asymmetry in the subjects is measured.  This is akin to an experimenter taking a reading of the resting pulse rate or resting blood pressure of the subjects before the experiment begins.  The purpose of this procedure is to categorize the subjects.  One method of categorization involves collating the subjects into three groups: left dominant, right dominant, and middle.

Once the subjects are thus categorized, the experiment begins and the subjects are presented with certain stimuli.  In trait asymmetry experiments, the stimuli tend to be somewhat ambiguous.  That is, they tend to be stimuli to which different kinds of people will typically react differently.  Eliciting different reactions is the point of such experiments.

These trait asymmetry experiments then conclude by correlating the different categories of reactions to the different categories of PFC asymmetry.  To the extent there is a significant correlation between the two, the experiment has revealed something about how PFC asymmetry relates to temperament — in other words, to mood.  Given that the title of this chapter is "Mood," observe that trait asymmetry experiments will play a particularly significant role in this chapter.

State asymmetry experiments are crucial to this chapter in another way.  Whereas trait asymmetry experiments look directly for how mood correlates with PFC asymmetry, state asymmetry experiments "drill down" on how specific emotions correlate with specific patterns of PFC asymmetry.

Accordingly, state asymmetry experiments tend to use fairly unambiguous stimuli.  The notion is that most everyone is expected to react in a similar way to these kinds of stimuli.  In these experiments, checking is typically done to ensure that the subjects did indeed experience a similar reaction.

These state asymmetry experiments then conclude by examining the patterns of PFC asymmetry exhibited by the subjects who reacted to the stimuli in the same way.  If that pattern is consistent across the subjects, then the experiments typically conclude that that particular reaction is mediated by that particular pattern of PFC asymmetry.  Typically, these experiments will announce that some particular emotion, or set of emotions, is processed by the left or right PFC. Then, collections of state asymmetry experiments, which collectively span many different kinds of emotions, can be reviewed to discern patterns.

In this way, studying collections of state asymmetry experiments is a "bottom up" way to approach the sort of conclusion that trait asymmetry experiments tend to pursue directly.  "Mood" is a very broad concept.  Making up this broad concept are many different kinds of specific emotions, including, for example, joy, enthusiasm, pride, sadness, fear, disgust, and anger.  Collections of certain of these specific emotions comprise a particular mood.  For example, joy, enthusiasm, and pride can be considered part of a "positive" or "up" mood.  Similarly, sadness, fear, and disgust can collectively describe a "negative" or "down" mood.[18]

So state asymmetry experiments are necessary for building up a theory of mood.  Then with this theory, trait asymmetry experiments can be designed to test the theory.  As time proceeds, later state asymmetry experiments may come along that have the effect of supporting or, alternatively, casting doubt on that theory.  This is the way in which state and trait asymmetry experiments interact.  The primary objective of this interplay is to arrive at theories that more closely approach truth.

Cognition versus Emotion

One final distinction to understand before we launch into the asymmetry research is the difference between cognition and emotion.  The discussion in the previous section on the differences and relationship between trait and state PFC asymmetry experiments concerned only studies of emotion, not of cognition.

Cognition is defined as "[t]he mental process of knowing, including aspects such as awareness, perception, reasoning, and judgment."[19]  A dictionary definition of "emotion" is "[t]he part of the consciousness that involves feeling; sensibility."  So, roughly speaking, cognition can be thought of as "thinking;" emotion as "feeling."  Convention has it that these two mental processes are very different, and quite independent.  Later, this chapter points to research showing that this distinction is not so sharp.  But, for now, the distinction is useful to draw.

Up until the past ten years or so, neuroscientific study of the brain was dominated by the study of cognition, not of emotion.  It is only in the last decade that the neuroscientific study of emotion has emerged in force.  In 1995, Mr. Davidson co-authored a paper entitled "Affective neuroscience: the emergence of a discipline."[20]  "Affective neuroscience" is the name Mr. Davidson coined to refer to the study of emotion in the brain.  As that paper explained:

As research progresses in this area, it is clear that the study of emotion, just like cognition, will require a dissection of emotional processes into more elementary operations, such as the perception of emotional information and the production of expressive behavior and autonomic activity, whose neural substrates can be better understood.[21]

This passage highlights the comparative importance of both emotion and cognition to the understanding of humans.  The passage goes further and highlights two key branches of research into emotion: perception ("perception of emotional information") and experience ("production of expressive behavior ...").

With respect to the study of cognition, trait and state PFC asymmetry are approached in a particular way.  In fact, the terms "trait asymmetry" and "state asymmetry" typically do not even appear in the field of cognitive neuroscience.  Even so, the concepts have meaning in this field.

Concerning trait asymmetry in cognition, a relevant question is: How does the thinking of people of different baseline PFC asymmetry differ?  In other words, how do righties think?  How do lefties think? As it turns out, these questions are too vague at present.  "Thinking" is such a broad notion, incorporating so many sub-processes that, at present, overarching theories of how PFC asymmetry corresponds to thinking are still a matter more of speculation, than of solid research.

So in the field of cognitive neuroscience, state asymmetry experiments dominate.  In the neuroscientific study of emotion, we saw that state asymmetry experiments proceed by attempting to elicit the same emotional reaction from all subjects.  In cognitive neuroscience, the game is slightly different.

Cognitive neuroscience makes use of the nature of human perception.  Humans have two eyes and two ears that perceive light and sound, respectively.  As it turns out, the right eye sends the light it perceives to the left side of the brain; the left eye, to the right side.  The same "crossing over" dynamic holds for hearing, as well as for touch.[22]

So when a cognitive neuroscientist wants to study, say, how the left PFC responds to certain visual stimuli, all that researcher needs to do is to display the visual stimuli to the right eye of the subject, and keep the stimuli hidden from the subject's left eye. Then, the performance of the subject in response to the stimuli can be attributed to the left PFC.[23]

Where the study of cognition and emotion merge is in the use of modern technology like fMRI and PET.  With these technologies, the researchers — both of cognition and emotion — present the stimuli and simply "roll the tape" (i.e. record what happens in the PFC of the subjects).  As a later section explains, this class of recent research is revealing the seamless connection between emotion and cognition.


With this introduction, we are now ready to dive into the asymmetry research.  The next section deals with asymmetry in cognition.  The remaining sections of this chapter look at asymmetry in emotion.  The latter form of asymmetry plays the greatest role in mood and personality.  That's why this chapter focuses mostly on it.  But some aspects of mood and personality are cognitive in nature.  So that's the main reason why cognition is addressed below. 

But there's another reason for addressing cognition.  Later, this chapter argues that if the correct theory of mood asymmetry concerns optimism and pessimism, then this theory, and the prevailing theory of cognitive asymmetry, seem to mesh nicely.   In fact, a plausible argument exists that the grand, classical subject of cognition may turn out to be "merely" a special case in the long-ignored, much-maligned subject of emotion.

But first things first.  Let's begin with asymmetry in cognition.

Asymmetry in Cognition

Among the most respected authorities on the matter of cognitive asymmetry has been Joseph Hellige of Harvard University.  Mr. Hellige's 1993 book, Hemispheric Asymmetry: What's Right and What's Left, has been recognized as of one the leading works in the field.

In Hemispheric Asymmetry, Mr. Hellige makes clear that the differences between left and right PFC with respect to cognition are not gross or large.  Popular misconceptions of this asymmetry hold that the left PFC is rational; the right, artistic, and creative.[24] Other misconceptions hold that thinking is a left function; feeling, a right function.[25]  Or that the left is for language, whereas the right is for orienting ourselves in space.

Mr. Hellige makes clear that these simplistic notions are incorrect.  There are indeed "lefties" and "righties" in terms of activation.  But regarding cognition, the functional distinction between the two sides is far more narrow and subtle then these broad, crude notions.  As we'll see in the next section, the left versus right distinction in emotion is more dramatic.  But in cognition, asymmetry is a delicate matter.[26]

Mr. Hellige's hypothesis on the cognitive difference between left and right PFC has been described by Robert Ornstein as the difference between "text" and "context".[27]  Toward understanding what this means, first recall the analogy some pages back concerning the submarine, the captain, and the periscope.  At any moment in time, the captain's consciousness of what is going on at the surface is limited to what he sees through the periscope (ignore for now, other sensory tools like radar).  He could move the periscope to look at something else.  But that would be a different moment in time.  At any particular moment in time, the captain's awareness is limited to his conscious field.

Now, taking this notion of the conscious field, observe that text and context are relative notions.  They are relative to the conscious field.  As the conscious field changes, so do text and context. 

For example, consider any sentence in this paragraph.  The text of that sentence includes every individual word.  The context of that sentence concerns its overall meaning.  Next, consider a particular word from that sentence.  Now your conscious field has shifted from the whole sentence, to a single word in the sentence.  With this as the present scope of your conscious field, text is now the individual letters of the word, and context is the whole word.  Another way to characterize this difference is "local" versus "global", or "parts" versus "whole".[28]

This contemporary understanding of cognitive asymmetry evolved out of decades of cognitive experiments that teased out one narrow difference or another. In other words, this larger understanding has been achieved by stitching together numerous state asymmetry findings. As noted earlier, trait asymmetry experiments are presently infeasible in cognitive neuroscience.

Representative examples of these state asymmetry findings include experiments on: language, vision, working memory, and apprehension. Starting with language, Mr. Hellige explains that:

[T]he left hemisphere seems dominant for the production of overt speech, for the perception of phonetic information, for using syntactic information, and for certain aspects of semantic analysis. However, the right seems dominant for certain other aspects of language, including the use of pragmatic aspects of language (e.g., narrative-level linguistic information) and the use of intonation and prosody to communicate emotional tone.[29]

Concerning the left side, "overt speech" refers to the use of words to communicate.  In speaking, this involves translating thoughts into vocal sounds that symbolize meaning.  "Phonetic" information involves the translation of written symbols into sounds (e.g. "B" translates to the sound "buh").  Syntax refers to the rules of grammar for constructing a proper sentence.  Semantics involves the translation of symbols into meaning. 

Relatively speaking, these are narrow features of language. In contrast, the right hemisphere is used for the broader forms of communication.  The "pragmatic" aspects of language include the "context in which an utterance occurs".[30]  For example, patients with damage to the right PFC don't "get" jokes or metaphors.

Early on in the previous century, famous author F. Scott Fitzgerald once said: "The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function."  He couldn't have known it at the time, but what he was talking about was a function of the right hemisphere.  The key to "getting" jokes, metaphors, and poetry is "the ability to hold two opposed ideas in mind at the same time [while] still retain[ing] the ability to function".  So, in his famous statement, Mr. Fitzgerald was insulting extreme "lefties" who have trouble "getting it". 

The same sentiment was expressed by a more recent Pulitzer Prize winning novelist, Richard Ford. Mr. Ford expressed his antipathy toward "literalists".[31]  Putting together these sentiments of Messrs. Fitzgerald and Ford with a sentiment expressed by award-winning author Andrew Solomon ("If we all felt up and great ... all the time, we could get more done and might have a happier time on earth, but that idea is creepy and terrifying"), a pattern begins to emerge.  It would seem that skilled writers don't much care for the lefties.  Well, the lefties will get their revenge later in the discussion on health.

The other language advantage of the right hemisphere concerns "the use of intonation and prosody".  Intonation refers to tone of voice; prosody, to accent.  The importance of these non-verbal aspects of language cannot be overstated.  Consider the phrase "I love you".  Using different tones of voice and different accents, a speaker of such words could convey meanings (sincerity, ambivalence, certainty, threat, compassion, sarcasm, etc.) so different as to be opposite. Tone and accent are the arrows in the quiver of "passive aggression".

This discussion on language illuminates the difference between text and context; parts versus whole.  The same dichotomy is evident for vision. Mr. Hellige describes a most elegant experiment involving brain-damaged subjects.[32]  The subjects were presented with pictures like the following:

z     z

zz   zz

z z z z

z  z  z

z     z

z     z

Hierarchical Stimulus

In the experiment, the patients with right hemisphere damage noticed the Zs, but not the M.  Patients with left hemisphere damage showed the opposite pattern, noticing the M, but not the Zs.  Thus, the left hemisphere notices parts; the right notices the whole.

Another instance of this finding is provided in an experiment that serves as the backbone of Mr. Ornstein's book.[33]  This experiment used subjects with brain damage to their right hemisphere.  These subjects were presented with a Norman Rockwell painting and asked to state what this picture was about.[34] 

The nature of the painting was such that, in order to correctly interpret it, the subjects needed to simultaneously hold in mind several of the picture's visual cues ("context"). Merely stepping through each cue individually ("text") would lead to an incorrect result.  All of the subjects proposed incorrect results. So the experiment corroborated that right PFC is necessary for capturing the visual "whole" or "context".

One interesting dynamic coming out of Mr. Ornstein's book is the confidence with which some of the brain damaged subjects seemed to offer their incorrect interpretations. In other words, it doesn't seem as though the subjects demurred from issuing conclusion about the picture on the grounds of lacking sufficient information upon which to base a conclusions. Instead, they just barged ahead confidently, albeit incorrectly, with partial information.

If this dynamic brings to mind current American business and political leadership, perhaps this is no accident. I'm not necessarily suggesting brain damage informs our nation's highest offices. I'm merely suggesting that our present business and political culture may skew appreciably left in terms of PFC activation. In my opinion, this skew becomes even more apparent in the discussion on PFC asymmetry in emotion.

The preceding discussion on language and vision corresponds to cognitive research of the past few decades. But with the recent advent of PET and fMRI, the field has branched out. Researchers are now asking how the PFC manifests text versus content; parts versus whole. Earlier research revealed the existence of this asymmetry; current research is explaining the mechanisms by which this asymmetry operates.

This recent research looks at the different way in which the parts are represented and accessed in working memory, from the way the whole is represented and accessed.  It seems that the parts are represented in an abstract, analytical way.  Moreover they seem to be accessed sequentially, not all at once.  In contrast, the whole seems to be represented more directly, and less as a tight abstraction.  Moreover, the whole seems to be accessed all at once.

These conclusions arise from a number of recent studies. One of these found that the working memory in the left PFC retains information for a longer duration than does right-sided working memory.[35] In explaining this results, that study suggested that the two sides use different strategies for representing stimuli.[36]  Specifically, the proposed strategy for the left PFC is analytic-based representations; for the right, it is image-based representations.

A later study[37] expanded on this proposal. In this study, various sounds were presented to the subjects.  Subjects were asked either to judge the loudness of the sound, or to decide whether the sound could be verbally described.  Later, their memory was tested.  The study found that the left PFC was used for representing the "verbal description" information about the sounds, while the right PFC was used for the "loudness" description of the sounds.

This study highlighted the difference between an actual sound, and word that represents the sound.  Consider the difference between the sound of music, and the musical score for that sound.  The musical score is the set of rules for generating the sound.  It's not the sound itself.  The amount of information needed for storing sounds, versus just storing the musical score for the sounds, is great.  Many musical scores can be stored in the space needed for storing the sounds for only one of the scores.  This illustrates the difference between an analytic-based representation of a stimulus; and a direct visual or auditory representation of the stimulus.

Researchers reason[38] that this difference between an abstraction of a stimulus, and the sensations of the stimulus, explains the durational differences between left and right working memory. Storage in working memory requires activation of the neurons in that area.  And activation entails the expenditure of energy.  So the more space in working memory that is required for storing a piece of information, the more energy that is required for this storage.  Since it appears that right working memory stores the sensations of stimuli, while the left stores abstractions of them, it follows that, for any particular stimulus, the right requires more storage space than the left.  The right would consume more energy in doing so.  Accordingly, the right would tire before the left.

This provocative finding suggests that, all things being equal, left PFC dominant people tend to abstractly analyze a stimulus more effectively than do right PFC dominant people. At the same time, the finding suggests that the latter tend to "get" the "whole story" of the stimulus, while the former are getting only pieces of it.

The final cognitive attribute addressed in this section concerns apprehension. This book uses the term "apprehension" to reference the ability to perceive negative emotions in others.  In neuroscience, we see that this is a function of the right PFC. Actually, as Mr. Hellige explains, "the right hemisphere ... seems to be superior to the left for the perception of both positive and negative emotions."[39]  Included in this notion is the superiority of the right PFC in "identifying the emotional tone of spoken material."[40]

In the previous paragraph, Mr. Hellige emphasized the word "perception." He did that because, as the next section shows, while perception of emotion, both positive and negative, may be primarily a right PFC function, the experience of positive and negative emotion is not.

Beyond apprehension of emotion in others, Mr. Hellige shows that the right PFC is also more attentive and alert.  The right PFC seems to be dominant for taking advantage of warning signals.[41]  Moreover, the right PFC seems to be able to maintain continuous attention on a stimulus for a longer period than can the left PFC.[42] Put simply, the right PFC is the "watchdog" of consciousness.

Apprehension of emotion, and wary attention, serve as an ideal stepping off point for the next subject: asymmetry in emotion.

Asymmetry in Emotion

As of this writing, emotion and cognition were still being treated as relatively separate domains within the field of neuroscience.  However, recent studies are indicating that the two domains seamlessly integrate, and perhaps even overlap.[43]  However, for the purposes of the present discussion, this book will continue to consider the two domains as separate.

As mentioned, Mr. Davidson has coined the phrase "affective neuroscience" to refer to the study of emotion in the brain.  "Affect" and "emotion" are synonyms. Perhaps Mr. Davidson chose the name "affective neuroscience" for the domain was because it fits nicely beside the name "cognitive neuroscience".  Although the name "emotional neuroscience" also lines up nicely, it has the undesirable side effect of conjuring up the "Far Side" image of emotionally distraught brain scientists trapped in hysteria.

Concerning the study of affect, Mr. Davidson has written: "Differences among people in affective style appear to be associated with temperament, personality, and vulnerability to psychopathology."[44]

Finally, deep within Chapter 5, we have come to the "beginning of the ending" of this book.  Here, with this statement, we have a leader in the field of neuroscience saying that human personality seems to be a function of the brain.

Specifically, Mr. Davidson mentions "affective style" and also "temperament, personality, and vulnerability to psychopathology."  "Affective style", as this chapter describes below, is a term Mr. Davidson uses to refer to certain patterns of brain functioning.  "Temperament" is a synonym for "mood" which is the title of this chapter. "Personality" is the subject of this entire book.  Accordingly, the theories of Mr. Davidson are central to this book.

 Since the processing of emotion is key to understanding how personality manifests in the brain, a useful next step is to ask: What is the purpose of emotion?  The answer coming out of neuroscience seems to be: "for decision-making".[45]  As Mr. Davidson explains: "Complex decisions ... such as who to marry, which job to take ... cannot be made solely on the basis of a cold calculus that involves the weighting of pros and cons in a formulaic prescription."[46]  Thus, effective decision-making "promote[s] adaptation."[47]

Note that the sort of decision-making being discussed here involves consciousness.  That is, at least some part of this decision-making process, which is driven by emotion, and which effects mood, is conscious.  This is distinct from emotional decision-making that is wholly unconscious.  This latter sort of decision-making will be discussed in the next chapter.

The preceding discussion indicates that emotion is used for decision-making, and that certain patterns of brain activity are associated with the experience of emotion.  We are now ready to look at asymmetry in emotion.  In other words, what sort of decision-making is processed by the left PFC, and what sort by the right PFC?

The current, generally accepted answer in neuroscience is that the left PFC mediates "approach;" the right PFC, "withdrawal".[48] Mr. Davidson explains:

[W]e suggest that left-sided PFC regions are particularly involved in approach-related appetitive goals. The instantiation of such goals, particularly in the face of strong alternative responses, requires left-sided PFC activation, and hypoactivation in these circuits has been linked to depression.  Right-sided PFC regions, alternatively, are hypothesized to be particularly important in behavioral inhibition and vigilant attention that often accompanies certain aversive emotional states and traits.  Whether right-sided PFC activation is a core feature underlying withdrawal behavior in general or behavioral inhibition and vigilant attention more specifically is a question to which we still do not have an adequate answer.[49]

This approach/withdrawal dichotomy is most evident in the studies of children. In a 2003 study, eighty-five 6-month-old babies were used as subjects.[50]  In the experiment, baseline resting PFC asymmetry was measured and the infants were divided into three groups: Left, Middle, and Right.  Then, the infants were seated in a high chair and subjected to a procedure called "stranger approach."  "During the stranger approach, a male stranger entered the room, slowly approached the infant with a neutral expression, and stared at the infant for up to 2 min."[51]  During this, EEG recordings were made of activity in the PFCs of the infants.  Also, the infants were observed and coded for signs of "facial fear, facial sadness, bodily fear, bodily sadness, vocal distress (crying), and escape behaviors."[52]

The researchers found that PFC asymmetry measured during the stranger approach predicted the sadness/fear response.  That is, those infants who exhibited extreme right PFC activity during the stranger approach were the ones to exhibit sadness/fear.  Infants who exhibited more middle or extreme left PFC activity during that task displayed much less sadness/fear.  This sadness/fear response, correlated with right PFC activity, is called the "withdrawal" response.

Interestingly, the researchers further found that resting baseline PFC asymmetry only weakly predicted the withdrawal response.  That is, just because an infant showed baseline extreme left PFC activity, that didn't necessarily mean he would not display a withdrawal response during stranger approach.  Conversely, infants showing baseline extreme right PFC activity did not necessarily exhibit withdrawal responses.  The researchers found only a "moderate" correlation between the baseline groups (Left, Middle, Right), and the stranger approach groups (Left, Middle, Right).  Put crudely, from baseline to the stressful stranger approach task, some lefties became righties or middies; some middies became lefties or righties; and some righties became lefties or middies. 

The subtle and complex relationship between trait (baseline) PFC asymmetry and state (task-based) PFC asymmetry is emerging in the most recent studies like this one.[53]  Earlier studies that introduced the basic notion of PFC asymmetry in emotion — whether trait asymmetry or state asymmetry — tended to ignore this complex trait/state relationship.  So the older studies need to be read with particular care.

One older study by Mr. Davidson and his colleagues sheds further light on the approach/withdrawal dichotomy.  That 1993 experiment studied 381 children aged 31 months (i.e. a brigade of "terrible twos").[54]  In the experiment, two unfamiliar same-sex peers were escorted into a playroom with their mothers, and left in the room for 25 minutes.  During the experiment, the mothers sat on chairs and did not interact with their children.  "There were age-appropriate toys on the floor in the playroom, including a toy tunnel through which the children could crawl."  At minute ten, a robot was brought into the room. The robot talked and walked toward the children.  At minute 20, a stranger entered the room with a tray of "very interesting-looking toys" and invited the children to play with the toys.

The children were observed and measured according to how much they engaged with the stimuli in the playroom.  From these measurements, three groups were formed: extreme engagement, extreme disengagement, and middle.  Actually, Mr. Davidson used the terms "uninhibited" and "inhibited" instead of "extreme engagement" and "extreme disengagement".  But I use the latter terms here because they are more narrowly descriptive, and less loaded.

Note this 1993 "terrible twos" experiment concerned only individual play, rather than group play.  Furthermore, the activities seemed to focus predominantly on outward exploration and adventure, rather than on, say, quiet observation, creativity, nurturing, or aggression.  It's not difficult to imagine how the experiment could be changed to study these latter sorts of attributes.  In those cases, "inhibited" and "uninhibited" could take on very different meanings.  This is why I stick to the narrow notion of "engagement".

Returning to the 1993 study, Mr. Davidson and his colleagues found that, on average, the extreme engagement group spent only 9 seconds out of the 25 minutes sticking close to mother.  During the rest of the 24 minutes and 51 seconds, these little dynamos raced around the room engaging with anything and everything they could get their eager little fingers on.

At the other extreme, the extreme disengagement group spent, on average, more than 19 of the 25 minutes clinging to mother, and engaging with few, if any, of the "fun" stimuli.  The middle group fell in between the two extremes on these measures.

At age 38 months (6 months later), these children were measured via EEG for PFC asymmetry.  The researchers found that the three groups differed according to baseline activation of left PFC. That is, the extreme engagement group showed, on average, high baseline activation of left PFC; the extreme disengagement group showed, on average, low baseline activation of left PFC. The middle group showed, on average, mid-level baseline activation of left PFC.

Notice in this 1993 "terrible twos" experiment, no state asymmetry readings were taken.  That is, while the children were racing around the room, or clinging onto mother, no EEG readings were taken.  Only trait asymmetry readings were taken after the fact.  Furthermore, these trait asymmetry readings were reported only in terms of group averages.

Despite these limitations, the 1993 study did hint strongly at something: namely, it suggested that the left PFC is associated with fun-loving adventurism.  Thus, we've seen that the left/right approach/withdrawal dichotomy seems to have something to do with fun-loving adventurism versus expressed sadness and fear, at least in infants and in toddlers.  But extrapolating these findings to adults is tricky business.  Adults are much more complex and ambiguous in the way they express fear, sadness, and joy.

For example, a 2002 experiment studied women and their eating habits.[55]  Starting with 55 women, two groups — restrained eaters, and unrestrained eaters — were formed based on answers to a questionnaire.  "[T]he restrained  eater is especially sensitive to anxiety-provoking situations; when exposed to such situations, restrained eaters typically overeat."[56]  The rest of the time, these people tend to be "chronic dieters" (i.e. restrained eaters).

This experiment measured baseline (trait) PFC asymmetry.  The study found that, on average, the restrained eaters were right PFC dominant, whereas the unrestrained eaters showed no dominant PFC asymmetry pattern.  In other words, some of the unstrained eaters were right PFC dominant.  Specifically, of the 23 restrained eaters in the study, only 18 showed a pattern of right PFC dominance (i.e. 5 were not right PFC dominant).

This study indicates that the right PFC plays some role in the behavior of restrained eating.  But that role is not altogether clear.  After all, restrained eating behavior in an adult is a much more complex and ambiguous activity than is crying for a 6-month-old baby.  For this reason, I believe studies that look to "sociability" versus "shyness" as the hallmarks of the approach versus withdrawal, left PFC versus right, dichotomy[57] reach too far.  "Sociability" and "shyness" are notions that seem too vague and ambiguous to be useful here.

The same criticism applies to "approach" and "withdrawal".  As some researchers have noted, an instance of behavioral withdrawal could be interpreted very differently depending on the imputed motivation of the subject.  For example, the subject could be engaging in instinctive flight (i.e. action without thought), cognitive pessimism (i.e. "That thing will hurt me" = withdrawal from danger), or cognitive optimism (i.e. "I can run to safety" = approach toward safety).[58]  These are radically different interpretations of the same observed withdrawal. Thus, to understand "approach" and "withdrawal", we need to drill deeper.

Valence versus Sensitivity

In the literature, there have been two different models for clarifying the vague approach/withdrawal diathesis: (1) valence; and (2) sensitivity.  According to the valence model, the dichotomy is between positive and negative affect.  Under the sensitivity model, it is between optimism and pessimism.

Until 2004, Mr. Davidson had been the principal proponent of the valence model.  Earlier, this chapter noted that Mr. Davidson associates affective style with personality generally, and with mood specifically.  About affective style, Mr. Davidson's statement as recently as 2003 was: "[A]ffective style [is defined] as valence-specific features of emotional reactivity and affective responding."[59]  Since mood and personality are trait-like attributes, we can conclude that, as of 2003, Mr. Davidson saw PFC asymmetry as a matter of emotional valence.

Consistent with this 2003 view of Mr. Davidson is the PANAS scale.  "PANAS" stands for "Positive and Negative Affect Schedule".  The PANAS is a questionnaire that probes the mental state of the subject.  This questionnaire asks subjects to rate themselves on positive emotions like "interested", "excited", and "strong", and on negative emotions like "distressed", "guilty", and "hostile". Up through 2003, Mr. Davidson had consistently reported that the positive schedule of PANAS correlates with left PFC, and the negative schedule with right PFC.

However, in 2004, Mr. Davidson reversed course and now seems to have abandoned the valence model. He writes: "[Eddie Harmon-Jones] argues that [the PFC asymmetry research is] consistent with the approach-withdrawal framework rather than with a valence-based model for frontal EEG asymmetry. More than 20 years ago, I argued the same point ... ."[60]

As of 2004, Mr. Harmon-Jones was a junior colleague of Mr. Davidson's at the University of Wisconsin-Madison. Since 1998, Mr. Harmon-Jones has been publishing EEG studies dealing with the emotion of anger.  The present conclusion of these studies is that optimistic anger is a function of the left PFC.[61] Optimistic anger is anger coupled with the belief that the anger will help achieve a positive result. Furthermore, Mr. Harmon-Jones has shown that the degree of activity in left PFC correlates with the degree of aggression in optimistic anger.[62]

The reason why optimistic anger precludes the valence model is that anger is generally considered "a negatively-valenced but approach-related emotion"[63].  Consistent with this view, the PANAS counts "hostility" and "irritability" — two low-ebb anger emotions — as negative emotions. Mr. Harmon-Jones' work casts doubt on the degree of correlation between PANAS and PFC asymmetry.

With the valence model having been discarded, the only remaining model for explaining approach/withdrawal is the sensitivity model. The principal authors of that model are Charles Carver and Teri White.  Messrs. Carver and White are the authors of the BAS/BIS Scales.[64]  Like the PANAS, the BIS/BAS Scales is a questionnaire that seeks to mine the emotional profile of the subject.

Specifically BIS, which stands for "Behavioral Inhibition System", "is sensitive to signals of punishment, nonreward, and novelty."[65] BAS, which stands for "Behavioral Activation System" or "Behavioral Approach System", "is said to be sensitive to signals of reward, nonpunishment, and escape from punishment."  Defined this way, BIS is simply pessimism ("a tendency ... to take the gloomiest possible view"[66]) and BAS, optimism ("a tendency ... to dwell on the most hopeful aspects of a situation"[67])

Moreover, as will be seen below, BAS/BIS is predominantly social in nature.  That is, the sorts of rewards and punishments upon which BAS/BIS focuses are social rewards and social punishments.  This is in contrast with non-social sensory rewards (e.g. taste of sugar) and non-social sensory punishments (e.g. pin prick).

An important distinction between the discarded valence model and the sensitivity model is along the dimension of objectivity/subjectivity.  It appears that the valence model relies upon objective notions of positive and negative. In contrast, subjectivity plays a significant role in the sensitivity model.  Although stimuli in such experiments are often deemed as objective "rewards" or "punishments", room is left within this model for subjective appraisals of the stimuli.  In other words, for the same stimuli, one individual may approach it as a reward, while another may withdraw from it as a punishment.  Even more complex, the same individual can approach the same stimulus optimistically one moment, and pessimistically later.  This complexity in the sensitivity model impacts experimental design.

Conceptually, the sensitivity model includes valence as a special case.  The above discussion on optimistic anger reduces the valence model to optimistic positive emotions, and pessimistic negative emotions.  Since optimistic states of mind are frequently accompanied by positive emotions, and since pessimistic states of mind are most often accompanied by negative emotions, the valence theory serves as a useful, albeit crude and limited, approximation for describing the left/right PFC dichotomy. 

That the sensitivity model was more apt than the valence model was evident even before Mr. Harmon-Jones' work on anger.  As early as 1997, Mr. Davidson had discovered that the BAS/BIS Scales (sensitivity model) are more accurate than PANAS (valence model):

More recently with Sutton ... we showed that scores on a self-report measure designed to operationalize Gray's concepts of Behavioral Inhibition and Behavioral Activation (the BIS/BAS scales ...) were even more strongly predicted by electrophysiological measures of prefrontal asymmetry than were scores on the PANAS scale ... .[68]

Still, it was only in 2004 that Mr. Davidson was ready to altogether let go of the valence model. Now, with that model gone, we are left only with optimism and pessimism.


In his most recent review of the field, Mr. Davidson asserts that sensitivity to reward (optimism) is a left PFC function, and sensitivity to punishment (pessimism), a right PFC function.[69] If this assertion is correct, one might reason that BAS/BIS should be reliably reflected in the PFC because it attempts to measure optimism/pessimism.  As of this writing, this is the case for BAS, but BIS is still an open question.

Three independent studies, including one by Mr. Davidson and one by Mr. Harmon-Jones, have confirmed that BAS is mediated by the left PFC.[70] In defining BAS, Messrs. Carver and White expanded the conception of optimism to embody three distinct forms: Reward Responsiveness, Drive, and Fun Seeking.

BAS Reward Responsiveness

When I get something I want, I feel excited and energized.

When I'm doing well at something, I love to keep at it.

When good things happen to me, it affects me strongly.

It would excite me to win a contest.

When I see an opportunity for something I like, I get excited right away.

BAS Drive

When I want something, I usually go all-out to get it.

I go out of my way to get things I want.

If I see a chance to get something I want, I move on it right away.

When I go after something I use a "no holds barred" approach.

BAS Fun Seeking

I will often do things for no other reason than that they might be fun.

I crave excitement and new sensations.

I'm always willing to try something new if I think it will be fun.

I often act on the spur of the moment.

Figure 10.# BAS Scale

In our current popular culture, these three forms of optimism have their own pet idioms:

      Reward Responsiveness: "Winning isn't everything it's the only thing"; "Just win, baby"; "Nice guys finish last"

      Drive: "Go for it!"; "It's better to ask forgiveness than permission"

      Fun Seeking: "You only live once"; "Been there, done that"; "Are you having fun yet?"

Beyond popular culture, these three forms of optimism found their way into a ten-year-old study that Mr. Davidson co-authored.[71]  This study focused on a certain type of person known as a "repressor".  

Repressors ... demonstrate a regulatory style that would appear to inhibit the perception of threat and the experience of negative affect and to promote the maintenance of self-esteem.  For example, such individuals demonstrate (a) a self-serving attributional style ...; (b) a self-serving hindsight bias ...; (c) impaired memory for negative self-relevant feedback ... and negatively toned autobiographical events ...; (d) attentional avoidance of threatening cues unless provoked by strong self-presentational concerns that may render heightened attention adaptive ...; (e) relative inability to consciously perceive negative affective stimuli under specific conditions ...; and (f) more classically defined defense mechanisms characterized by the inhibition of interpersonal conflict and ambivalent or negative emotions and by selective accentuation of the positive ... .[72]

The foregoing definition highlights the "dark side" of blind optimism.  In fact, if we consider the dark side of the three forms of BAS optimism, we arrive at elements (d), (e), and (f) of the repressor definition.

Reward Responsiveness is all about winning.  Where the desire to win becomes obsessive, the interest in being perceived by others as a winner soon becomes paramount.  In this case, "self-presentational" concerns become central.

Drive is all about relentless pushing.  When the pushing becomes obsessive, maintenance of it requires the pusher to be unaware of "negative stimuli" that could interfere with the pushing.

Fun Seeking is all about the fun.  Where the pursuit of fun becomes maniacal, the pursuer has no time for "ambivalent or negative emotions" and needs to "accentuat[e] the positive".

Just as the BAS studies found that BAS is a left PFC function, this 1994 study of repressors found that the repressors are left PFC dominant.[73] Interestingly, the subjects of this study were 90 women.  Of these 90 women, somewhat less than 25% of them were designated as repressors.  This study alone suggests that while something on the order of 75% of women may be from Venus, a good percentage of women are every bit as Martian as most men.


Although it seems settled science that BAS (optimism) is mediated by the left PFC, BIS (pessimism) as a right PFC function is still controversial.  In a 1997 paper he co-authored, Mr. Davidson reported that BIS is correlated with right PFC.[74] But a 1997 study co-authored by Mr. Harmon-Jones found no such correlation.[75]  And a 2003 study reported only "a weak statistical trend toward relatively greater right frontal activity corresponding to higher BIS scores in the midfrontal region only."[76]

What is going on here?  The researchers suggest various explanations for this conflict concerning BIS.  I suspect the conflict has to do with the nature of the BIS scale itself.  Read again the BAS scale shown in Figure 10.##.  In our current culture, affirmative responses to none of those questions suggests anything pejorative about the respondent.  For example, being a person who feels free to "go for it" is not generally frowned upon in current American popular culture.

Now consider the BIS scale:


If I think something unpleasant is going to happen I usually get pretty "worked up".

I worry about making mistakes.

Criticism or scolding hurts me quite a bit.

I feel pretty worried or upset when I think or know somebody is angry at me.

Even if something bad is about to happen is about to happen to me, I rarely experience fear or nervousness. [reverse scored]

I feel worried when I think I have done poorly at something.

I have very few fears compared to my friends. [reverse scored]

Figure 10.# BAS Scale

The BIS scale asks subjects to admit to the experimenters that they are fearful, anxious, hysterical, hypersensitive, and/or guilt-ridden.  I submit that, in our popular culture, these are all pejorative characterizations.  Given that all of the BIS experiments mentioned above studied college students, the question becomes: What is the likelihood that college-age kids would freely admit to pejorative, "judgmental" characterizations of themselves?

I suspect that the answer to this question depends heavily on the experimental environment.  Did the experimenters create an atmosphere of loving acceptance, in which subjects could feel comfortable in admitting to personal vulnerabilities?  Obviously, there's no way of determining that from reading these cold studies.

So if we are going to conclude that BIS (pessimism) is mediated by the right PFC, we will need to look elsewhere for evidence.  Fortunately, the emergence of PET and fMRI have obviated the need to rely on self-report measures like the BAS/BIS Scales.

Measuring Optimism and Pessimism Directly

In his 2003 review of the field, Mr. Davidson reported that: "[A] a left-sided medial region of the orbitofrontal cortex (OFC) appears particularly responsive to rewards whereas a lateral right-sided region appears particularly responsive to punishments (O'Doherty at al., 2001)."[77]

This 2001 O'Doherty study comes out of the Wellcome Department of Imaging Neuroscience, of the Institute of Neurology, in London, England.  This paper is one among a pair of recent papers published by the Wellcome Department concerning how the PFC relates to monetary reward and punishment.  In addition, Mr. Davidson's lab has also generated studies in this specific area.

Using money as the "carrot and stick," these studies focused on a particular type of social reward and punishment.  Money is an abstract notion having no meaning outside of social life.  In contrast, taste and touch are physical sensations that do not depend on social life.  So studies that test for glucose taste rewards[78] or prickly touch punishments[79] test physical rewards and punishments.  Those latter kinds of studies do not test for the type of optimism and pessimism with which the sensitivity theory of emotional asymmetry deals.  So this section focuses on the experiments that study responses to money.

In the money reward/punishment experiments, the subjects were typically instructed to play a game in which their choices determine a monetary reward or punishment.   In an old (1992) EEG study co-authored by Mr. Davidson,[80] subjects were presented with "reward trials" and "punishment trials".  In these trials, the subject were forewarned about the type of trial they were in.  After a delay, they were presented with an opportunity to gain the reward (reward trial), or to avoid the punishment (punishment trial).

The study recorded PFC activity during the anticipation phase of the trials (i.e. the delay period between the warning, and the action opportunity).  The study found that left PFC was activated during the anticipation phase of the reward trials; and right PFC was activated during the anticipation phase of the punishment trials.[81]

Recent work in this area has come out of the Wellcome Department in London.  Over the past few years, that lab has sought to isolate the particular sub-regions of the PFC that play a part in reward/punishment recognition, anticipation, experience, and subsequent response.  The main focus of this research has concerned the identification of these sub-regions, rather than identifying hemispheric asymmetry in reward/punishment processing.  Nevertheless, the detailed findings of this research reveal just such asymmetry.

In the 2001 study cited by Mr. Davidson,[82] the London researchers instructed subjects to play a game on a computer.  In this game, subjects faced four possible outcomes: win big money, win small money, lose small money, or lose big money.  The choices made by the subjects during the game determined the outcomes.  The game required the subjects to be constantly deciding on the best choice to make to maximize gain and minimize loss. fMRI was used to record the responses of the subjects to their gains and losses.

The study found that the greater the monetary gain, the greater the left PFC activation.[83]  Conversely, the greater the monetary loss, the greater the right PFC activation.[84]  Interestingly, the study did not find this asymmetry for any particular level of gain or loss. 

This last point highlights the subtle subjective nature of optimism and pessimism.  A particular outcome might be considered as a reward by one subject, but as a punishment by another.  But a second outcome of greater monetary value than the first would be considered relatively rewarding by all subjects.  This may explain why PFC asymmetry was found for relative rewards and punishments, but not for absolute ones.[85]

These studies suggest that the left PFC biases our attention to socially rewarding cues, and processes our response to receiving such rewards; and that the right PFC attends to and processes socially punishing cues and outcomes.  This suggestion dovetails nicely with one of the fundamental purposes of emotions as described by another researcher out of the Wellcome Department in London:

An evolutionary perspective on emotion suggests that environmental events of value should be susceptible to prefrontal perceptual processing.  One means of achieving this is emotion enhancing attention, leading to increased detection of emotional events.[86]

Under this formulation, "approach versus withdrawal" as a description of left versus right PFC asymmetry in emotion acquires specific meaning.  "Approach" becomes "sensitivity toward social reward", where "reward" is defined subjectively.  Similarly, "withdrawal" becomes "sensitivity toward social punishment", where "punishment" is a subjective notion.

Finally, note that the optimism/pessimism dichotomy provides a straightforward answer to a question posed by cognitive asymmetry expert Joseph Hellige.  Mr. Hellige observed that whereas our right hemisphere perceives both positive and negative emotion in others, both hemispheres are active in our personal experience of emotion, with the left mediating positive emotions, the right, negative ones.  Concerning this observation, Mr. Hellige noted:  "It is instructive to consider how these different findings might be reconciled."[87]

A simple answer for Mr. Hellige comes from the optimism/pessimism dichotomy.  Since the pessimist attends to social punishment and the optimist to social reward, and since certain social punishments can kill us but no social reward can, it falls to the pessimist to be well aware of the emotions of others.  Sensing reward, the optimist feels free to train his attention on himself.  But these are not conscious decisions.  They are, instead, more or less unconscious biases.  This explains the role of the right PFC in detecting the emotions of others.  On the question of experienced emotional valence, it takes little imagination to consider why an optimist expecting and acknowledging reward might, on average, feel more positively and less negatively than a pessimist expecting and acknowledging punishment.

Thus optimism/pessimism not only falls out of direct recent research.  It goes further and elegantly answers open questions in the field of cognitive neuroscience. Sounds like we have winner here.

One Caveat

Although the case for optimism/pessimism appears quite persuasive, it is presently not airtight. This is because it has some shortcomings. The following describes the case presented by this chapter, and its shortcomings:

                        Messrs. Carver and White claim that BAS/BIS (optimism/pessimism) results from two different brain systems; however, to my knowledge, they haven't attempted to verify their claim directly (i.e. through EEG, fMRI, and/or PET).

                        Mr. Davidson claims direct validation for the theory that optimism/pessimism, and BAS/BIS, are mediated by left PFC/right PFC, respectively; however, there are some suggestions in his recent work that he hasn't quite let go of the notion of positive/negative traits (valence), even though this notion seems insupportable.[88]

                        Mr. Harmon-Jones, and other researchers, claim direct validation for the theory that BAS (optimism) is mediated by left PFC; however, they claim they have been unable to verify BIS (pessimism) as a right PFC function.

                        Recent research conducted by the Wellcome Department in London directly demonstrated optimism and pessimism as functions of the left and right PFC, respectively; however, these researchers were not looking specifically for optimism/pessimism; moreover, to the extent their research unintentionally unveiled optimism/pessimism, it did so only in the specific context of money gain/loss.

As of this writing, these shortcomings precluded the sensitivity model (optimism/pessimism) of PFC asymmetry from being considered conclusively verified as fundamental.  However, that model is currently the only one proposed in the field for clarifying the inherent ambiguity of the approach/withdrawal diathesis.  Moreover, no one has yet disproved this model. Thus, even in the absence of a single, peer-reviewed, widely-accepted line of research supporting this model, it stands as the one that is most promising and altogether probable.

Physical Health and PFC Asymmetry

The remainder of this chapter looks at how PFC asymmetry relates to physical health, genetics, and plasticity (the ability to change).  We begin with physical health. This section concludes that PFC asymmetry has a significant impact on physical health.

One critical factor in physical health is the functioning of our immune systems.[89]  In the body, the immune system wards off toxins and assists in the repair of damage.

In conflict with the immune system is the stress system.  The stress system serves to mobilize the body to respond to a stressful situation.  The immune system and the stress system relate in a roughly zero-sum way.  Normally, the immune system is functioning.  But if the stress system in invoked, the immune system is dampened, allowing the stress system to draw upon the body's available energy stores.

One principal mechanism of the immune system is natural killer ("NK") cell activity.  NK cells are generated by the immune system, and delivered to the sites of toxic invasion.  There, the NK cells attack and destroy the invasive toxins.  In general, presence of NK cells in the blood is an indication of a functioning immune system.

One principal marker of the stress system is the hormone cortisol.  Faced with a stressful situation, the brain signals the adrenal glands ## to secrete cortisol.  This hormone acts as a bodily alarm bell, rousing other stress system components into action, and instructing the immune system to "stand down".

In the short term, in the face of momentary stressful situations, the stress system has a dramatic positive impact on physical health.  In a truly stressful situation — one in which our lives are threatened — prompt and effective functioning of our stress systems can mean the difference between life and death for us.

However, where our stress systems act continually, over long stretches of time, physical health is impacted negatively.  One main reason for this is that over these long stretches, the immune system is compromised.  This provides ample opportunity for toxins to enter the body and do their damage, untroubled by the quieted immune system.

For testing the effect of PFC asymmetry on physical health, neuroscientists have examined cortisol response and NK cell activity.  Cortisol levels are measured in the saliva or the blood; NK cell activity is measured in the blood.  For examining cortisol, studies tend to use infants as subjects because the association of PFC asymmetry with cortisol is much clearer for infants than it is for adults.

Studies of both human babies[90] and monkey babies[91] have found the same thing: PFC "righties" exhibit higher levels of cortisol activity than "lefties".  What is unclear is which hemisphere is responsible for the heightened cortisol activity.  In both the human and monkey experiment, middle asymmetry groups were studied, as well as lefties and righties.   The monkey experiment found that the middle group aligned with the righties; the human experiment found the opposite: namely, the middle group lined up with the lefties.  So the former experiment suggests that the activation of right PFC mediates the release of cortisol; the latter suggests this release has something to do with hypoactivation (de-activation) of the left PFC. Whatever the case, one thing is clear: among all the asymmetry groups, the immune systems of righties are the most compromised.

Consistent with this finding, a 1999 study reported on NK cell activity in college age students.[92]  That study found that "[s]ubjects with greater relative left-sided anterior [prefrontal] activation ... showed higher levels of NK activity."[93] This study showed directly that the immune systems of lefties tend to be stronger than those of righties.

Mr. Davidson participated in each of the cortisol response and NK cell activity studies described above.  In his 2003 review paper, he described his most recent extension of this work:

Very recently, we ... have extended this work to include measures of in vivo immune function. ... [W]e administered an influenza vaccine and measured antibody titers in response to the vaccine at several internals following vaccination.  We found that subjects with greater left-sided prefrontal activation at both baseline and in response to influenza vaccine, suggesting more robust immunity in response to vaccination.[94]

Earlier, this chapter noted the antipathy that some great literary authors express toward lefties, and their shortcomings with respect to metaphorical and holistic awareness.  There I said the lefties would "have their revenge".  This section describes that revenge.  Righty says: "Lefty, you are one ignorant literalist."  Lefty responds: "Make that blissfully ignorant, righty.  And when flu season comes around next, feel free to borrow my Kleenex and chicken soup, because you'll be needing them, and I won't."

Genetics and PFC Asymmetry

Given the dramatic differences between left and right PFC on the matters of optimism, pessimism, awareness, and physical health, two questions arise: Does trait PFC asymmetry have a genetic basis?  Can trait PFC asymmetry be altered through behavior?  This section looks at the former question; the next section, at the latter.

The human baby experiment mentioned in the previous section sheds some light on the first question.  This 2003 experiment is the same as the one discussed earlier in this chapter.  That study examined PFC asymmetry in 6-month old babies, both at baseline, and during the stressful "stranger approach".

That this study recorded reliable PFC asymmetry in 6-month old babies suggests a genetic origin of this asymmetry.  Of course, it's entirely possible that environmental events ("nurture") during the six months outside of the womb, or during the previous nine months inside, caused the asymmetry.  But it seems at least equally likely that genes ("nature") played the dominant role.

One aspect of this study appears, on initial review, to suggest the environmental answer.  The study included five mono-zygotic twins.  This means these twins came from the same egg and sperm; thus they share the same genes.  For only one of these five sets, the two twins fell into the same PFC asymmetry group (e.g. right-dominant) during the stressful event ("stranger approach").  That means the twins of the other four sets presumably exhibited different PFC asymmetry.  If PFC asymmetry was a genetic characteristic, one might have expected every set to fall into the same asymmetry group.

However, this expectation is not well founded.  Earlier, this chapter explained that individuals frequently exhibit variability in patterns of PFC asymmetry.  Most days, for example, a certain person might exhibit right-sided dominance.  But on certain other days, the same person might fall into the middle group.  This dynamic is the exception to "test-retest" reliability.

The other relevant dynamic here is the complex relationship between trait and state asymmetry.  This baby experiment found that trait asymmetry (i.e. asymmetry measured at baseline) was only weakly correlated with state asymmetry (i.e. asymmetry measured during stranger approach).  Indeed, the researchers found that cortisol activity was associated only with trait right-dominant PFC asymmetry, not state.[95] As noted earlier, as of this writing, no systematic studies have been conducted explaining this complex trait-state relationship, let alone discovering the causes of variability.

It is entirely possible that the variability observed in PFC asymmetry between test and retest, and between trait and state, is caused by environmental factors.  Perhaps one the twins was just having a difficult day and that explains the difference in asymmetric pattern. If this is so, then the findings of this baby experiment on mono-zygotic twins are inconclusive on the question of genetic causation.

As of this writing, the most that can be said about a genetic origin of trait PFC asymmetry is a solid "maybe".

Plasticity and PFC Asymmetry

If it turns out that patterns of trait PFC asymmetry have genetic origin, critical questions arise: Are lefties condemned to blissful ignorance?  Are righties sentenced to a lifetime of sickly, miserable awareness?  The answer coming out of neuroscience is a resounding "No!".  The brain is much more plastic than that.

Plasticity refers to the ability to develop new pathways among the neurons, and even in some cases, to generate new neurons ("neurogenesis").  Put simply, plasticity concerns change in the operation of the brain.  Among the beneficial aspects of plasticity is the process of "transforming emotional habits."[96]  Since, as we've seen, PFC asymmetry plays a large role in emotional habits, the discussion here on brain plasticity focuses on the alteration in patterns of PFC asymmetry.

As of this writing, neuroscience seems rather confident of the claim that transforming PFC asymmetry is possible.  Since the field currently focuses more on the downside of extreme right PFC dominance, and less on that of extreme left PFC dominance, work in this area concerns the amelioration of right dominance. 

But although the field seems confident of the possibility of transformation, it is far less confident about the methods for accomplishing that.  Undoubtedly, the pharmaceutical industry is watching this area closely.  One need not strain one's imagination to picture adverts for an "Optimism Pill"" on weekday morning television programming (some might say we are already there).

For now, at least, neuroscience seems more interested in publishing studies concerning behavioral, rather than pharmaceutical, methods for achieving this transformation.  Methods currently identified include biofeedback training, massage therapy, and music therapy.  Concerning negative transformation, child abuse has been implicated.  In addition, although meditation does not currently seem to implicate PFC asymmetry, it does appear to bear on a pattern of immune response that correlates with left PFC dominance.

Concerning biofeedback training, Mr. Harmon-Jones and colleagues published a study in 2001 in which the subjects were 18 women aged 18-38.[97]  The women were fitted with EEG caps, and were asked to listen to the sound of continuous tones.  Among these tones was a high-pitched, "reward" tone.  The subjects were instructed to "try to make the high tone stay on."[98]  Evidently, this involved "focus[ing] on the high tone."

In the experiment, the 18 women were divided into two groups: LEFT and RIGHT.  For the LEFT group, the experiment nudged the subjects toward left PFC activation.  That is, the high tone would sound only if PFC asymmetry in the subject had shifted sufficiently toward the left.  For RIGHT subjects, the dynamic was opposite.

The study found that the training worked.  Specifically, subjects trained LEFT did exhibit greater relative left PFC asymmetry; and subjects trained RIGHT showed the opposite transformation. Interestingly, the results for "right" training were better than those for the "left" training.  For these 18 subjects at least, it appeared that it was easier to train greater pessimism into them than greater optimism.  Moreover, although training-directed changes to PFC asymmetry were observed in this experiment, there was no suggestion that these changes were persistent. 

Short-term changes in PFC asymmetry have also been observed in response to massage therapy and music therapy.[99]  Although short term, the researchers reported both therapies to be "extremely effective".  Concerning music therapy, they explained:

[A]fter only 20 minutes of music (in this case, rock music), 10 of the 12 depressed mothers showed an attenuation of right frontal EEG activation, moving toward symmetry or toward left frontal EEG activation ... . The two adolescents whose EEG patterns did not change claimed that they did not enjoy rock music.  When their favorite music (classical) was played, they too experienced a shift toward symmetry.[100]

Tragically, more pervasive changes — perhaps permanent ones — have been recorded as a result of child abuse.  Harvard researcher Martin Teicher is a leader in this field.  He has written:

Our research team used this technique [EEG coherence] in 1997 to compare 15 healthy volunteers with 15 child  and adolescent psychiatric patients who had a confirmed history of intense physical or sexual abuse. ... The right hemispheres of abused patients had developed as much as the right hemispheres of the control subjects, but their left hemispheres lagged substantially behind.  This anomalous result showed up regardless of the patient's primary diagnosis.  ... [T]he effect extended throughout the entire left hemisphere.[101]

Observe that this study concerned, not so much PFC asymmetry in activity, but rather PFC asymmetry in development.  Presumably, developmental asymmetries are even more difficult to ameliorate.

The final method discussed here concerns the practice of meditation.  In his 2003 review paper, Mr. Davidson discussed his recent study on meditation:

We found that subjects in the meditation group showed an increase in left-sided anterior activation whereas subjects in the wait-list control group showed a change in the opposite direction. ... [Also], we found that subjects in the meditation group showed a larger increase in antibody titers to the [administered] influenza vaccine compared with the controls ... .[102]

In referring to "anterior activation" in the preceding paragraph, Mr. Davidson was not referring to PFC activation. As he reports:

We predicted that we would find significant changes in prefrontal as well as central electrode locations.  It is unclear why our most consistent findings were observed at the central leads. ... The fact that there was no significant increase in dispositional positive affect in the meditation group may be related to the failure to detect increases in left prefrontal activation.

As of this writing, what can be said about meditation is that this practice is correlated with a greater immune response, but with neither greater left PFC asymmetry nor greater optimism.  This suggests the existence of a third process that mediates the immune response.  Left PFC activation, and the effects of meditation, seem to be two different ways of impacting this third process. They appear to be two different roads heading to the same place.

Earlier this chapter discussed the "revenge of the lefties". Namely, the advantage that lefties have over righties is that, all things being equal, the former naturally have stronger immune responses than the latter.  However, for evening the score, this latest research provides righties with a secret weapon: meditation. That is, even if the state of insufferable, perpetual happiness remains a remote possibility for righties, the state of physical health is easily attained. (But if it takes optimism to see this, or even to want it, perhaps the term "easily attained" is a tad optimistic here.)

Cognition: Merely A Special Case of Emotion?

The final subject of this chapter concerns some speculation of mine about the proper relationship between cognitive neuroscience and affective neuroscience.  Earlier, this chapter explained that the former has long dominated the study of the brain.  Affective neuroscience is the recently emerged, poor neglected sibling of its celebrated cognitive twin.

But if the conclusion of this chapter is correct ... namely, that PFC asymmetry in emotion is about optimism/pessimism ... then this imbalanced relationship may soon witness a reversal.  This is because the diatheses coming out of cognitive neuroscience ... parts/whole, text/context, local/global, analytic/holistic ... seem merely to be special cases of optimism/pessimism.  The argument for this is a short one:

According to evolutionary psychology, humans can be thought of as survival machines.  If this is so, then the role of PFC asymmetry in modulating optimism/pessimism emerges as fundamental.  This is because optimism concerns sensitivity toward survival enhancing prospects, while pessimism concerns toward threats to survival. These are conscious processes.

Now threats to survival can come from any direction.  If the organism fails to spot even just one such threat, and the threat turns out to be fatal, this failure will prove to be the organism's last.  So pessimism requires attention to be attuned relatively broadly.  Pessimism needs to be sensitive to everyone and everything, just in case.

Optimism, in contrast, has no such need. If optimism fails to notice a survival-enhancing prospect, the cost of that failure is merely a missed opportunity. In general, missed opportunities rarely prove fatal.  So, optimism is free to focus attention only upon the most promising of prospects.  In fact, generally speaking, the more concentrated the focus, the greater the chance optimism has of realizing the promising prospect.  So optimism is best served with attention attuned relatively narrowly.

This suggests that the appropriate level of sensitivity of optimism/pessimism is narrow/broad.  But this dichotomy aptly describes parts/whole, text/context, local/global, analytic/holistic.

Indeed, this discussion highlights the dual nature of optimism/pessimism.  There is the "hot", emotional form of it; and there also is the "cold", rational, emotionless form of it.[103]  So optimism/pessimism seems to straddle both emotion and cognition.

Hence, it just may turn out that cognitive neuroscience, which has long called the shots in brain research, will soon find itself taking marching orders from affective neuroscience, the domain in which optimism/pessimism emerges most clearly.


After a lengthy and somewhat dense discussion, Chapter 3 concluded by pulling optimism/pessimism, and aware/unaware fear out of the Enneagram.  There, this book advised that the sausage (the conclusion), was more important than the sausage-making process (the discussion).  Indeed, that lengthy discussion was included primarily so that Enneagram experts could review and criticize the basis for the conclusion.

This chapter has followed a similar dynamic.  The lengthy and somewhat dense discussion of this chapter concludes that optimism/pessimism is mediated by left/right PFC, respectively, and that humans fall somewhere along the continuum from left to right PFC dominance. This discussion is directed primarily toward the neuroscience community. It is up to that community to decide whether the reasoning that leads up to that conclusion is sound. But for now, just note that, at a minimum, the neuroscience literature adequately supports optimism/pessimism as a key description of PFC asymmetry. 

With that, we are done with mood. We are now ready to enter the chamber of fear.


[1] LeDoux/SS at 177-78.

[2] LeDoux/SS at 174-99.

[3] LeDoux/SS at 179.

[4] LeDoux/SS at 191.

[5] See LeDoux/SS at 192.

[6] Davidson (2004)

[7] Davidson (2003)

[8] Davidson (2003) at 658.

[9] Davidson (2003) at 658.

[10] Davidson (2003) at 658.

[11] Coan/Allen (2003) at 7.

[12] Coan/Allen (2003) at 7-8.

[13] Davidson (2003) at 658.

[14] Coan/Allen (2003) at 9-10.

[15] Coan/Allen (2003) at 26-34 calls for research in this area.

[16] Coan/Allen (2003) at 3.

[17] Coan/Allen (2003) at 5: "[T]rait frontal EGG asymmetry will refer to asymmetries that are consistent intra-individually across time ... . State frontal EEG asymmetries, by contrast, may be thought of as those that are responsive to specific environmental conditions."

[18] Perhaps you're thinking: "What about 'anger'?  Where does it fit in?"  If so, collect two points for your keen powers of observation, and read on.


[20] Davidson/Sutton (1995)

[21] Davidson/Sutton (1995) at 217.

[22] ## cite

[23] The researcher can draw this conclusion because the functioning of the two hemispheres is relatively independent.  Hellige (1993) at 112.  That is, instead of the right and left PFC furiously passing signals back and forth in performing some function, each side more or less does its work by itself, and then the final result of processing of each side is seamlessly unified in human consciousness.  Hellige (1993) at ##.

[24] Hellige (1993) at 1.

[25] Hellige (1993) at 2.

[26] ## Hellige believes single fundamental dichotomy may be futile — but for gross tasks — maybe not for subprocesses; Hellige (1993) at 63-6

[27] Ornstein (1997) at ##.

[28] Hellige (1993) at ##.

[29] Hellige (1993) at 61-62.

[30] Hellige (1993) at 38-39.

[31] ## cite "The Sportswriter"

[32] Hellige (1993) at 74-75.

[33] Ornstein (1997).

[34] Ornstein (1997) at 140.

[35] Courtney (1998) at 1824.

[36] Courtney (1998) at 1824.

[37] Optiz (2000)

[38] See Courtney (1998) at 1824.

[39] Hellige (1993) at 50.

[40] Hellige (1993) at 51.

[41] Helige (1993) at 104-105.

[42] Hellige (1993) at 105.

[43] See, e.g., Gray (2002).

[44] Davidson (1998) at 308.

[45] Bechara et al (2000); see also Dolan (2002) at 1191 (emotion is for memory, motivation, and belief fixation).

[46] Davidson (2003) at 655.

[47] Davidson (2003) at 655.

[48] Davidson (2003) at 655 ("approach and withdrawal [are] two of the core dimensions").


[50] Buss (2003)

[51] Buss (2003) at 12.

[52] Buss (2003) at 12.

[53] See Coan/Allen (2003) at 29-34, 35-36 for an in-depth discussion of possible models for trait/state associations.

[54] Davidson: Cerebral Asymmetry, Emotion, and Affective Style at 378-382.

[55] Silva at al (2002)

[56] Silva at al (2002) at 677.

[57] Schmidt; Coan/Allen (2003) at 14.

[58] Coan/Allen (2) (2003) at 107.

[59] Davidson (2003) at 657

[60] Davidson (2004) at 224.

[61] Harmon-Jones/Allen (1998); Harmon-Jones et al (2000).

[62] Harmon-Jones et al (2000); see also Coan/Allen (2003) at 26 which cites this work.

[63] Coan/Allen (2003) at 13.  See also Coan.Allen (2003) at 4 ("The valence model is, obviously, unable to accommodate these recent anger findings, as anger is a negatively valenced emotion characterized by left frontal activation").

[64] Carver/White (1994).

[65] Carver/White (1994) at 319.



[68] Davidson (2003) at 658.

[69] Davidson (2003) at 656; see also Harmon-Jones/Sigelman (2000) at 7 (Mr. Harmon-Jones speculating that "hope" is the core dynamic of left PFC function).

[70] Coan/Allen (2003); Harmon-Jones (1997); Davidson/Sutton (1997).

[71] Tomarken/Davidson (1994)

[72] Tomarken/Davidson (1994) at 340.

[73] Tomarken/Davidson (1994) at 345.

[74] Sutton/Davidson (1997).

[75] Harmon-Jones/Allen (1997) at 162.

[76] Coan/Allen (2003) at 112.

[77] Davidson (2003) at 656.

[78] See, e.g., O'Doherty at al. (2002).

[79] See, e.g., Rolls at al. (2003).

[80] Sobotka (1992)

[81] Sobotka (1992) at 245.

[82] O'Doherty (2001)

[83] O'Doherty (2001) at 96.

[84] O'Doherty (2001) at 97-98.

[85] See also O'Doherty et al. (2003) (delving deeper into the various aspects of monetary reward and punishment processing).

[86] Dolan (2002) at 1191.

[87] Hellige (1993) at 52.

[88] ##explain

[89] ## cite for the immune and stress system discussion

[90] Buss (2003)

[91] Kalin (1998)

[92] Davidson/Coe (1999)

[93] Davidson/Coe (1999) at 106.

[94] Davidson (2003) at 660.

[95] Buss (2003) at 17.

[96] Davidson (2003) at 655.

[97] Allen/Harmon-Jones (2001).

[98] Allen/Harmon-Jones (2001) at 687.

[99] Field (1998)

[100] Field (1998) at 1308.

[101] Teicher (2002).

[102]   Davidson/Kabat (2003) at 662-663.

[103] A 1999 study seemed to have teased out this latter form of pessimism. Garavan (1999). In the study, the subjects were presented with a stream of letters and asked to make "a button response whenever certain letters (X or Y) were presented."  The extra wrinkle in the experiment was that subjects were asked to inhibit this response if an X followed an X (separated by at least 30 other letters), or a Y followed a Y (separated by at least 30 other letters).  This study, using fMRI, found that response inhibition was strongly associated with right PFC activation.