Functional Neuroanatomy and Brain Imaging
of Personality and its Disorders
C. Robert Cloninger
People differ markedly in their personality traits, but there is a stable organizational structure of the brain circuits regulating personality traits. In this chapter, a clinical model of personality will be described that corresponds well with DSM-IV categories of personality disorder and also with available data about the functional neuroanatomy of temperament and character. Temperament refers to the emotional aspects of personality that are regulated by subdivisions of the limbic system and centrally integrated in the hypothalamus. Character refers to the higher cognitive aspects of personality that are regulated by subdivisions of the thalamoneocortical system and centrally integrated in the frontal cortex. In this chapter, quantifiable personality traits will be related to a model of their functional neuroanatomy derived from classical and comparative neuroanatomy, along with supporting clinical data and experimental findings from brain imaging.
Temperament will be considered in terms of four dimensions of personality that are related to specific subdivisions of the limbic system. These traits are called 'harm avoidance' (anxiety-proneness versus libido, which is outgoing vigour and daring), 'novelty seeking' (exploratory impulsivity and irritable aggression versus frugality and stoicism), 'reward dependence' (social sensitivity attachment versus insensitivity and aloofness), and 'persistence' (industrious determination versus underachievement). A quantitative dimensional approach is used because this corresponds well with the functional neuroanatomy and because there is no evidence of discrete boundaries between traditional syndromes of personality disorder, as in DSM-IV. Instead, traditional syndromes emerge from the interaction of multiple dimensions, which give rise to specific configurations of temperament that distinguish developmentally stable subtypes (Cloninger, 1987; Cloninger and Svrakic, 2000).
Character will be considered in terms of higher cognitive functions that regulate an individual's goals and values. In other words, character refers to mental self-government, which includes executive, legislative, and judicial functions. These traits are called 'self-directedness' (purposeful and resourceful versus aimless and helpless in executive functions), 'cooperativeness' (helpful and principled versus hostile and opportunistic in legislative functions), and 'self-transcendence' (inventive and insightful versus unimaginative and undiscerning judgement) (Cloninger et al., 1993).
Each of the seven dimensions of personality considered here has been shown to have genetic variability that is unique from all the others, indicating that a model with seven dimensions is needed to describe the organization of human personality (Gillespie et al., in press). The correlations among the temperament dimensions are weak, as are the correlations among the character dimensions, but there are moderate non-linear relations between temperament and character. This is expected because character (i.e., higher cognitive functions) modulates temperament (i.e., emotionality).
A well-integrated character allows coherence of personality by modulation of emotional conflicts. In contrast, to the degree that character is not well integrated, emotional conflicts arise from the interplay of competing urges from various temperament dimensions. Likewise, the dynamics of complex adaptive systems oscillates between multiple states as a result of small contextual influences on dissociable circuits whenever such circuits are coordinated by a self-organized interplay, as has been suggested as a model of brain systems (Bressler and Kelso, 2001). Consequently, different dimensions of temperament are correlated with differences in the activity of specific brain circuitry that is partially overlapping but functionally dissociated when character is not coherent in its guidance.
All individuals with personality disorders have immature character development, particularly low self-directedness and low cooperativeness (Svrakic et al., 1993; Cloninger and Svrakic, 2000). Individuals with different DSM-IV categories of personality disorder are distinguished primarily by their temperament profiles. Descriptive labels for the personality subtypes defined by the temperament dimensions of novelty seeking, harm avoidance, and reward dependence are shown in Figure XXVI-6.1. For example, borderline personality disorder is associated with a temperament configuration comprising high novelty seeking, high harm avoidance, and low reward dependence. Individuals with antisocial personality disorder are also high in novelty seeking and low in reward dependence, but they are low in harm avoidance. Consequently, the brain images of individuals with antisocial and borderline personality disorder are similar in many, but not all, ways. Specifically, individuals with either borderline or antisocial personality disorder have brain images expected from low self-directedness (i.e., low activity in the medial prefrontal network) and low cooperativeness (i.e., low activity in the orbital prefrontal network) (Kuruoglu et al., 1996; Raine et al., 1998; London et al., 2000; Soloff et al., 2000). Likewise, both groups of individuals have brain images expected from high novelty seeking (e.g., high activity in the right insula) and low reward dependence (e.g., high activity in the right superior temporal gyrus) (Sugiura et al., 2000; Soloff et al., 2000). However, they differ in activity associated with harm avoidance, which is negatively correlated with activity in paralimbic regions such as the left parahippocampal gyrus, left orbitoinsular junction, and some neocortical regions such as the fusiform gyrus (Sugiura et al., 2000). Borderline subjects have the brain correlates of high harm avoidance (Herpertz et al., 2001), whereas antisocial subjects have the brain correlates of low harm avoidance (Raine et al., 1998; Schneider et al., 2000). Each of the dimensions of personality distinguished here has unique genetic variability and brain circuitry, so it is more informative to describe personality dimensionally than categorically. Studies of categorically defined groups of patients