The Emotional Foundations of Personality

by Kenneth L Davis

  The Big Five has emerged as perhaps the most scientific personality assessment model to this point, with no preconceptions about the underlying structure of mind. It was guided by the recognition that the language of personality should map well onto our linguistic use of adjectives to describe human traits. It let the chips fall in whatever way they would, based simply on allowing a statistical analysis of our language dictate what differences existed in our personalities. Clearly, our ANPS emerged from very different premises, namely, an understanding of the subcortical brain fundamentals of our emotional-affective natures. Pausing to consider these issues, before proceeding to the history of the field, we hope will help readers appreciate not only the dynamics of science in this contentious field but also possible lasting neurobiopsychological solutions. First, however, we backtrack a bit.

  TOWARD A THEORY OF PERSONALITY FOCUSED ON PRIMAL EMOTIONAL AFFECT

  As we describe in Chapter 15, there are significant connections between genetics and personality. In Chapter 3, we also summarize discussions by Darwin that animals exhibit emotional personalities similar to human personalities. In Chapters 4 and 5, we review how psychological and psychoanalytic pioneers at the beginning of the twentieth century engaged in active discussions about the relations between emotions and personality. But first, in this chapter we continue to make the case that modern affective neuroscience has, for the first time, provided a coherent, evidence-based view of mammalian minds as rooted in the value/survival-coding affective circuits situated at the very basement of the brain and mind. This kind of evidence makes a strong case that our personalities are firmly grounded in the ancestral roots of our biological bodies and brains. But what is it in our neurobiology that mediates the connections between our genes and our personality traits?

  As introduced in Chapter 1, Jaak Panksepp has outlined a dynamic view of human behavior that integrates an advanced understanding of mammalian brain anatomy, brain chemistry, and the diverse emotions that can be aroused by direct electrical stimulation of isolated deep brain structures. His explicit view is that various basic emotional/affective action systems, grounded in a rich evolutionary past, remain laid out in very similar ways in very ancient brain regions (Panksepp, 1982, 1998a). The more traditional behaviorist label for these brain emotion-generating systems is unconditional response systems. These emotional-action survival networks (or simply emotional instincts) constitute the evolutionary foundation of the “mind,” which is easier to talk about in many psychological terms (e.g., see Chapters 4 and 5) than to study neuroscientifically.

  The main theme of this book is that these primary emotional processes are the links between our evolutionary heritage and our current major personality characteristics. These emotional systems support subjective affective experiences that motivate and guide our actions, control learning and memory, and promote diverse cognitive activities. They may constitute the underlying basis for our individual personality traits (Davis, Panksepp, & Normansell, 2003; Davis & Panksepp, 2011).

  These are primary processes in that they do not need to be learned, although they control learning and reciprocally are refined by learning. That they have an evolutionary past is evident from the fact that all mammals share their instinctual-genetic foundations, and many of these primary processes are evident in birds and reptiles, perhaps even fish, although the degree of homology (genetic relatedness) remains ever more open as the degree of genetic relatedness diminishes. Still, it is the evolutionary relationship that we humans have especially with other mammals that has allowed for much of our current understanding of these emotional systems across species. Indeed, it is this same bodily relationship with mammals that has also allowed us to make giant strides in medical research and therapeutics based on working out details of underlying biological processes in animal models. This is also happening in psychiatry (Panksepp, 2005, 2016; Panksepp & Yovell, 2014).

  As noted already, there is substantial evidence for six primary emotional systems, which are closely linked to personality traits. Humans experience these emotional systems as motivating affects and potent urges to behave in particular ways. Subjectively, we experience three of these affects as negatives or aversive feelings in our lives: RAGE/Anger, FEAR/Anxiety, and PANIC/Sadness. We also subjectively experience three of them as positive, desirable, or pleasant feelings: SEEKING/Enthusiasm, CARE/Nurturance, and PLAY/Social Joy. (We leave out consideration of LUST/Eroticism for now.) Each of these six systems is adapted to promote our survival and responds characteristically to specific environmental situations and emergencies. When highly aroused, each of the six evokes powerful negative or positive affective feelings, and these systems promote distinctive instinctual reactions. We know that they feel like something to other animals as well as to humans, because deep brain simulation (DBS) of each of these brain systems is rewarding or punishing to animals. Our key hypothesis here is that the strengths and weaknesses in these six brain emotional affective systems, whether genetically dictated or developmentally refined, are the very foundations for the emotional traits that constitute our personalities. This beckoned us to get into the complex (at times chaotic) field of human personality.

  FROM PRIMAL EMOTIONAL AFFECTS TO PERSONALITY DIMENSIONS

  How can one measure the emotional strengths and weaknesses of people? How does one assess the activity in one’s emotion systems? It cannot yet be done clearly or directly by any physiological measures, even though there are many linkages. So how might clinicians and researchers appraise the personality relevant urges and affective tendencies of their subjects? At present, it can be done only by asking the right questions. How might research physicians evaluate the underlying primary emotional strengths and weaknesses of their patients, some seeking help for their addictions, to get to know them nomothetically (the study of universals that we share with others), not just idiographically (focusing on individual differences), to refine a diagnosis, prognosis, and treatment plan? Psychiatrists and clinical psychologists can benefit from psychometric scales that attempt to directly assess individuals’ emotional strengths and weaknesses. For instance, how might psychiatrists assess the extent to which addicts have experienced real changes in the emotional makeup during treatments for their addictions? No good tools yet existed. Thus, the first task we set for ourselves was to construct a psychological assessment tool that measured primal emotional traits. At first blush, one option seemed to be the Big Five personality model, which had emerged as a prominent factor-analytically derived model for parsing the human personality. Namely, it had revealed patterns of human personality based on assessment of a host of adjectives that had emerged in human language to describe temperamental differences among people. Furthermore, at present it is arguably the most widely accepted personality evaluation tool produced by psychological science.

  BRIEF INTRODUCTION TO THE BIG FIVE, AKA THE FIVE-FACTOR MODEL

  Unfortunately, the Big Five did not turn out to be a satisfactory model for assessing primary emotions. Why? Because it had no vision of the core emotional traits of human beings! Except for Emotional Stability, the Big Five did not deal directly with affects. Indeed, all negative (aversive) affects were lumped into the low pole of the Emotional Stability dimension, which is sometimes given the negative label neuroticism. From our perspective, there were no separate measures for the RAGE/Anger, FEAR, and PANIC/Sadness systems.

  The Big Five is the subject of later chapters as well, but here we provide a brief background. The Big Five model is the culmination of many years of psychometric research. It is the direct outcome of a statistical method called factor analysis and asserts that human personality variation can be described with five traits usually named Extraversion, Agreeableness, Conscientiousness, Emotional Stability, and Openness to Experience.

  There is an historical background that we do not focus on in this chapter, but there were many significant players. More than any other, it was Lewis Goldberg who brought the Big Five to its peak visibility (see Goldb
erg, 1990) by using factor analysis to place self-ratings of hundreds of English-language adjectives on the Big Five axes, including adjectives that seemed to “blend” more than one Big Five trait (Hofstee, de Raad, & Goldberg, 1992). He also provided researchers with a list of one hundred “marker” adjectives—twenty for each of the Big Five traits (Goldberg, 1992). Indeed, we used seventy of the clearest markers to measure the Big Five traits in our own personality research. Examples of these markers were “talkative” and “assertive” for Extraversion, “kind” and “helpful” for Agreeableness, “systematic” and “organized” for Conscientiousness, “irritable” and “moody” for low Emotional Stability, and “complex” and “imaginative” for Openness to Experience.

  However, despite all of Goldberg’s (and many other’s) fine theory-free work, all of the negative emotionality in human behavior was simply lumped into a single trait: emotional stability. Even though fear, anger, and sadness are distinct feelings for all humans, and even though fear, anger, and sadness, in their psychiatric extremes, are each associated with different psychological problems, with different recommended treatments, the most widely accepted psychometric personality model only offered a single personality dimension that lumped all three of these unique and powerful emotions into a single psychological trait.

  Because there were no satisfactory psychological tests, we decided to construct a tool to measure each of the six primary brain affects, which we call the Affective Neuroscience Personality Scales (ANPS). The ANPS was modeled after the State-Trait Personality Inventory (Spielberger, 1975, 1983) developed by Charles Spielberger (1927–2013). But the ANPS aimed to measure six emotional traits instead of Speilberger’s three (anger, anxiety, and curiosity, although Spielberger created many additional variants). The ANPS was designed as a self-report instrument that would tap into the six major affective-emotional tendencies hypothesized to underlie our personalities. It would be the first psychological assessment that would position an individual in affective space, based on our emerging knowledge of cross-mammalian core (primal-evolved) emotional systems.

  Tertiary Level of Emotion Processing

  At this point, a brief explanation is in order about levels of emotion processing (addressed in more detail in Chapter 5) and what we were actually attempting to measure with this new assessment. In Chapter 1 we discussed primary-level affective processes. We also described primary emotions as supporting learning, which in Chapter 5 we will refer to as secondary-level processing. This secondary adaptive emotional learning is based on the primary-level affective processes, thereby helping refine our responses to life events and create new survival responses to meet the challenges of an ever-changing world. The idea that humans, indeed all mammals, exhibit this secondary capacity to automatically learn from our affective experiences and adapt our behaviors to the specific demands of our environments makes us much less reflexive and more dynamic/flexible beings with multiple ways to handle specific environments.

  However, for humans—perhaps other animals as well, but especially humans, with our massive cerebral cortices and thereby capacity for speech—there is also a tertiary level of processing, based on thoughts and decision making, that is surely more complex than that found in any other species. (However, it may be that whales and dolphins—sea dwelling mammals—are exceptions.) This tertiary level is the level of cognitive processing or, more simply, just thinking. Because humans can talk about our experiences, including cognitive perceptions, thoughts, and memories, we are usually reporting our affective experiences to each other verbally, both in nonverbal tone and in semantic content. The importance of the primary, secondary, and tertiary levels is discussed in Chapter 5, but for now it is important to acknowledge that there are no clear scientific words to adequately describe what we feel at the primary level, and the data we sought to collect with our new assessment were language-based reports, namely, from the tertiary level of brain-mind processing.

  While we wrote test items that specifically asked about “feelings” as directly as possible, we recognized that the answers people would give us would inevitably include their reflections about their feelings rather than a pure “direct brain readout” of their primary affects. We avoided writing items like “I yell at other people less than my friends do” because this required the subject to make a logical comparison that would depend on how much those in his or her circle of acquaintances resorted to yelling. Rather, we tried to write items that could access the primary affective feeling level as directly as possible from a tertiary, language/thinking-based perspective. An example would be “I almost never yell at other people.” Our intention was to write items that would allow people to respond as directly as possible from the primary feelings that are embedded in their personal experiences of themselves.

  Getting Started

  After several iterations of writing items, testing students, and rewriting items, we finally had produced six clearly emotional personality scales (seven overall, counting Spirituality, which we discuss later). They had adequate reliabilities, which basically meant we had developed items for each scale that could be shown statistically to be consistently measuring the same underlying concept. We named the new assessment the Affective Neuroscience Personality Scales (ANPS) and started evaluating how well it worked in the real world, yielding several cycles of refinement of this new instrument as more and more data rolled in. In brief, college students were requested to complete the ANPS along with the short seventy-item test measuring the Big Five traits, as practical laboratory exercises in their ongoing psychology courses, first in Panksepp’s lab, followed by Ken Davis and Larry Normansell, and now many others across the world. For readers wishing to study the ANPS as revised in 2011, it is reprinted as Appendix A.

  Even though we did not consider the Big Five to adequately measure primary emotions, we still wanted to determine how these six affective ANPS traits related to the Big Five dimensions, especially because that had become the most widely accepted human personality model. In fact, we predicted from previous work that our PLAY scale would correlate with Big Five Extraversion, SEEKING would correlated with Openness, CARE would line up with Agreeableness, and RAGE/Anger, FEAR, and PANIC/Sadness would all clump together with Emotional Stability. Thus, we also hypothesized that, as indicated above, there would be significant cases where the Big Five may have blurred major distinctions between primary emotions.

  ORIGINAL ANPS RESEARCH FINDINGS

  The results showed that each of the six ANPS scales correlated most strongly, as predicted, with one of the Big Five dimensions, as shown in Table 2.1, which summarizes findings from our first ANPS publication (Davis et al., 2003). Among the positive affects, the PLAY scale lined up best with Extraversion, SEEKING linked clearly with Openness to Experience, and CARE correlated predominately with Agreeableness. Among the negative affects, RAGE (which we originally called ANGER), FEAR, and PANIC (which we originally called SADNESS) all correlated very highly with lower Emotional Stability. RAGE/Anger also showed a second relationship by lining up with low Agreeableness. The Big Five Conscientiousness scale correlated negatively with each of the three ANPS negative affects, although at a degree of magnitude lower than their correlations with Emotional Stability.

  Thus, we obtained evidence that the foundations of human personality may be laid in the old emotional parts of mammalian brains, and hence also the human brain, suggesting that the ancient instinctual action-based emotional-affective networks, originating in deep subcortical brain regions, may be major “drivers” of human personality dimensions. Even though we had left LUST out of our scale, we decided to include a tertiary Spirituality scale because of its general importance in human affairs, especially addiction treatment programs such as Alcoholics Anonymous, where clients are encouraged to develop more spiritual ways of dealing with life.

  Table 2.1. Correlation of Affective Neuroscience Personality Scales and Spirituality Scales with Big Five Personality Scales: Early Results


  Affective

  Neuroscience Personality Scales Big Five Personality Scales

  Extra-

  version Agreeableness Conscientiousness Emotional Stability Openness

  to Experience

  PLAY 0.46*** 0.29*** 0.00 0.12 0.13

  SEEKING 0.13 –0.01 –0.01 0.01 0.47***

  CARE 0.25** 0.50*** 0.12 –0.07 0.06

  FEAR –0.19* –0.17* –0.24** –0.75*** –0.05

  ANGER –0.04 –0.48*** –0.30*** –0.65*** –0.08

  SADNESS –0.21** –0.13 –0.30*** –0.68*** –0.00

  Spirituality 0.15 0.26*** 0.14 0.09 0.17*

  Student sample, n = 171 (50 males, 121 females). Adapted with permission from Davis et al. (2003).

  *p < 0.05; **p < 0.01; ***p < 0.001 (all two-tailed).

  Simple Scale-Level Interpretation

  The data supported the conclusion that each of the ANPS scales, which were based on extrapolations of the primary emotional systems of mammalian brains, were strongly related to one or another of the Big Five dimensions. RAGE/Anger was the only ANPS scale to be strongly associated with two Big Five dimensions. Parenthetically, we again note that the full capitalizations in the ANPS are official names for the primary-process emotional systems, but we will typically use the more vernacular designators, in lowercase, for discussing the fuller human-emotional complexities of personality. We do this because we have no direct neuroscientific way to monitor and probe the deep primal emotional systems of human brains, even though certain kinds of brain imaging (e.g., positron emission tomography) can visualize many of these systems (see Damasio, 1999; Damasio, Grabowski, Bechara, & Damasio, 2000). However, we assume that our language-based measures are actually making contact with the personality dynamics that are linked to (indeed, partly emerge from) such primal emotional strengths and weaknesses.