Gray's Motivational Model and Relations with Psychopathology

I have been focusing on empirically established relations between mood and personality which led to the hypothesis that mood forms an inherent core of at least two major dimensions of personality. I turn now to further theoretical efforts to understand and explain these relations. As mentioned, a number of writers have used Gray's three-factor motivational system to provide a theoretical framework for this domain. Gray (1982, 1987) has proposed a motivational model involving three biological systems: the behavioral inhibition system (BIS), the behavioral activation system (BAS), and the fight-flight system.

The BIS is a preparatory system, with coordinated aspects in response to conditioned aversive stimuli (both punishment and frustrative nonreward), novel stimuli, and innate fear stimuli. The BIS simultaneously inhibits ongoing behavior while increasing nonspecific arousal which, in turn, are together associated both with increased attention to environmental stimuli (vigilance and scanning for potential threats) and preparation of the organism for a vigorous response, if necessary. In essence, the BIS is a “stop, look, and listen” system that is activated in response to the possibility of danger.

P.181

Based primarily on pharmacological research using animals, Gray has argued that the BIS constitutes the neurophysiological substrate for anxiety. Fowles (1992, 1993) has elaborated on this view in terms of Barlow's (1988) two-factor theory of anxiety. Specifically, the BIS feature of continual threat appraisal is characteristic of what Barlow calls anticipatory anxiety, a key element in generalized anxiety disorder. Fowles argues further than the BIS feature of aroused inactivity, although not emphasized by Barlow, is consistent with his view of anticipatory anxiety as a preparation for stress and challenge.

The BIS may be considered as not only a state-respondent system, but also as a system that reflects individual temperament. From this perspective, individuals with a characteristically hyperactive BIS are oversensitive to negative stimuli, and they therefore frequently anticipate danger unnecessarily; such individuals are temperamentally tense, hypervigilant, and behaviorally inhibited. This description clearly resembles that of individuals high in neuroticism or negative emotionality, and Tellegen (1985) has linked the BIS with this personality dimension. With the connection established earlier in this chapter between negative affect—of which anxiety is a central emotion—and neuroticism/neg-ative emotionality, it appears that the circle is complete: The BIS appears to be the biological system underlying generalized (anticipatory) anxiety, negative affect, and neuroticism/negative emotionality. However, emphasizing that the BIS is a behavioral inhibition system and that neuroticism/negative emotionality are concerned primarily with subjective affective experience, Fowles (1992) has argued that the BIS should be linked with the disinhibition-constraint dimension, a viewpoint that I will discuss later.

By contrast, the BAS is an active or approach system that responds primarily to conditioned stimuli for reward but also for active avoidance of punishment (Gray, 1982). There is a positive hedonic tone to BAS activation (Fowles, 1993), an important characteristic in clarifying the link between the BAS and the personality dimension of extraversion, when reconceptualized as positive emotionality by Tellegen (1985). Similarly, Depue and colleagues (e.g., Depue, Luciana, Arbisi, Collins, & Leon, 1994) have argued persuasively that the BAS represents the same motivational system as their behavioral facilitation system (BFS), described as a positive emotional system that motivates exploratory behavior as a means of goal acquisition.

Like the BIS, the BAS has both state-response properties (in which signals for potential reward increase organismic activity) and also trait-like characteristics (in that there are temperamental differences in individuals' sensitivity to reward signals and, accordingly, in the strength of their behavioral and emotional responses). A number of research lines support these various connections. For example, the various component features of extraversion/positive emotionality covary in bipolar disorder (Depue & lacono, 1988; Depue, Krauss, & Spoont, 1987). Specifically, manic states are associated with euphoric mood, enhanced confidence and optimism, heightened energy and excitement seeking, and increased social/sexual interest. By contrast, depression has been described as a low positive affect state characterized by a sad, pessimistic mood, anhedonia, low energy level and sense of self-worth, and withdrawal from social and other stimuli (Clark & Watson, 1991b). To a lesser extent than in mood disordered patients, these characteristics covary in normal individuals as well (Watson & Clark, 1997). Finally, Depue et al. (1994) have recently demonstrated that extraversion/positive emotionality is related to individual differences in

P.182

the reactivity of the mesolimbic dopaminergic system, which Gray (1987) has identified as the neural basis of the BAS, and which has been implicated in bipolar disorder as well (Depue & lacono, 1988).

It is important to distinguish between the level or strength of the BFS/BAS and its variability. Depue et al. (1987) have argued that dysfunctional variabil-ity—due to poor regulation of the system—is what results in the appearance of bipolar disorder, whereas the strength or weakness of the system sets the level around which this variability occurs, which then determines whether the individual will manifest mostly manic episodes, mostly depressive episodes, or cycling from one to the other.

It is also important to note that as with the mood and personality dimensions, these underlying biological dimensions have at least a certain degree of independence. For example, to explain the high degree of comorbidity between anxiety and mood disorders, Clark & Watson (1991b) have proposed that both mood and anxiety disorders share a common NA/neuroticism factor. A great deal of evidence indicates that patients with these disorders—and patients with many other disorders as well—score high on common measures of this factor. Thus, BIS activation (if, indeed, that is the substrate for NA/neuroticism) may underlie a great deal of psychopathology. However, mood and anxiety disorders also have some distinguishing features; specifically, positive emotionality is negatively correlated with depression, but is unrelated to anxiety disorders (Clark& Watson, 1991b). That is, BAS activation and/or dysregulation is a depression-specific factor that distinguishes it from anxiety.

Clark and Watson (1991b) further proposed that heightened autonomic arousal was a specific marker of anxiety that distinguished it from depression. Thus, it is intriguing that the proposed mechanism of action for Gray's third fac-tor—the flight-fight system—is the sympathetic branch of the autonomic nervous system. This system activates a vigorous behavioral response to unconditioned aversive stimuli, including defensive aggression (fight) and escape (flight). Fowles (1993) has proposed that the flight-fight system corresponds to the second type of anxiety—alarm reaction or fear—in Barlow's (1988) two-anx-iety theory. According to Barlow, panic attacks occur when the alarm system is triggered inappropriately (i.e., without a clear unconditioned danger signal), thus constituting a false alarm.

In an insightful analysis of Gray's model in relation to psychopathology, Fowles (1993) reviewed Schalling's (1978) work on psychopathy and observed that she also distinguished between two types of anxiety. The first is psychic anxiety, which is identifiable as anxious apprehension (and, by extension, as negative emotionality and perhaps the BIS); the second is somatic anxiety, which bears some similarity to the alarm reaction and flight-fight system, being described as consisting of “autonomic disturbances, vague distress and panic attacks, and distractibility” (p. 88). Historically, psychopaths have been characterized as having low anxiety levels, but Schalling's data suggest that this may refer only to BIS-related psychic anxiety, whereas they may have high levels of fight-flight-related somatic anxiety. Of interest in this context is that Schalling (1978) found somatic anxiety to be positively correlated with impulsivity—an important component of the disinhibition dimension—in psychopaths.

This analysis raises the question of whether there is a third personality dimension corresponding to the fight-flight system, to parallel the BIS-NA, BASPA connections already discussed. Considering that the fight-flight dimension

P.183

is associated with the activation of vigorous behavior (e.g. aggression) and somatic anxiety (which is, in turn, associated with impulsivity), at first glance it would appear that the dimension of disinhibition, impulsive sensation seeking, or psychoticism versus conscientiousness/agreeableness of the five-factor approach discussed earlier is an obvious candidate. But there are some problems with this matching. First, we must reconcile the affective (strongly aroused anger/panic) component of the fight-flight system with the demonstrated lack of relation of disinhibition to mood. In this regard it is important to note that the data only indicate that disinhibition has no relation to trait levels of the two higher order mood dimensions, whereas a specific relation between disinhibition and hostility—yet another identified characteristic of psychopaths, and consistent with the “fight” component of the domain—is frequently reported (e.g., Watson & Clark, 1992).

While these data provide some support for the linkage, they are admittedly not as satisfactory as data showing that disinhibition also is related specifically to fear/panic. However, certain data suggest that if disinhibition has any relation at all to fear/panic, it is probably a negative rather than a positive correlation. For example, in his classic work on the lack of anxiety in psychopaths, Lykken (1957) used an anxiety measure now called harm-avoidance (Tellegen, 1985), which assesses a specific facet of the disinhibition-constraint dimension that has also been labeled physical-danger anxiety (conceptually and empirically distinct from NA-related psychic anxiety). Subsequent work also has confirmed that psychopaths tend to be electrodermally underaroused (Fowles, 1993), indicating a negative relation with fearfulness. In light of these data, it is problematic to link disinhibition positively with the fight-flight dimension. However, Fowles (1987) has suggested a way out of the dilemma, based on distinguishing trait from state anxiety. Perhaps disinhibited individuals, although they are temperamentally non-anxious (i.e., low psychic anxiety), actually experience more state (i.e., somatic) anxiety, precisely because they frequently engage in risky behaviors. Although perhaps not completely satisfactory, this analysis at least permits a reasonable consideration of the flight-fight system as the biological substrate for disinhibition.

Personality Disorder

Fundamentals of Personality Disorder

Turning aside for the moment from mood, personality, and their biological bases, I turn now to the third element in the chapter title—personality disorder. I first provide some background information for readers unfamiliar with this domain and then discuss the role of mood in defining personality disorder. I then link a recently proposed theoretical model of basic dimensions underlying personality disorder with the biopsychological temperamental dimensions already discussed.

Definitions of personality disorder typically are based largely on traits. For example, the text section in the Diagnostic and statistical manual of mental dis orders (4th ed.) on personality disorders begins by defining traits as “enduring patterns of perceiving, relating to, and thinking about the environment and oneself that are exhibited in a wide range of social and personal contexts” (DSM-JV;

P.184

American Psychiatric Association, 1994, p. 630). This definition is quite consistent with those offered in the psychological literature on normal range personality. DSM-IV then proceeds to specify further how personality traits must be manifested before they can be identified as constituting a personality disorder. First, the trait pattern must deviate markedly from the expectations of the individual's culture and be manifested in two or more of the following areas: cognition, affectivity, interpersonal functioning, and impulse control. A number of relevant affective parameters—range, intensity, lability, and appropriateness of emotional response—are specifically mentioned. In addition, the other general criteria emphasize that the defining traits must be stable and pervasive; moreover, in order to be viewed as the basis for a disorder, the traits also must be inflexible and maladaptive, causing either distress or social or occupational dysfunction. These general diagnostic criteria for personality disorders, which must be met to assign any Axis II diagnosis, are listed in DSM-IV for the first time and clarify both the trait-based nature of personality disorder and also the ways in which the trait manifestations must be deviant and maladaptive in order to be considered disordered.

There are 10 official personality disorder diagnoses in DSM-IV, and two provisional diagnoses “in need of further study” provided in an appendix. The personality disorders are grouped into three clusters based on certain descriptive similarities. It is acknowledged that “the clustering system… has serious limitations and has not been consistently validated” (APA, 1994, p. 630), but the organization is helpful to gain an initial understanding of these disorders. The Cluster A diagnoses of paranoid, schizoid, and schizotypal personality disorder are characterized as “odd and eccentric”; those of Cluster B—antisocial, borderline, narcissistic, and histrionic personality disorder—as “dramatic, emotional, or erratic”; and the avoidant, dependent, and obsessive-compulsive personality disorders of Cluster C as “anxious and fearful.”

DSM-IV introduces the criteria for each specific personality disorder with a general statement of their defining characteristics, followed by seven to nine more specific criteria, a subset of which must be manifested as exemplars of the defining characteristics in order to receive the diagnosis. For example, antisocial personality disorder is defined as “A pervasive pattern of disregard for and violation of the rights of others” (APA, 1994, p. 649), which must be fulfilled by manifesting three or more of the following specific criteria: repeated illegal acts, deceitfulness, impulsivity, irritability and aggressiveness, reckless disregard for safety, consistent irresponsibility, or lack of remorse.

From the viewpoint of social cost, personality disorders are quite problematic. By definition, they are associated with personal distress and/or social-oc-cupational dysfunction. Personality disordered patients are also considered unpleasant to interact with and difficult to treat (Vaillant, 1987). A recent literature review (Ruegg & Frances, 1995) reported associations between personality disorder and an extensive range of social ills from child abuse and neglect, homelessness, and underemployment to increased usage of medical care, malpractice suits, and disruption of treatment settings.

Evidence indicates that personality disorders are highly comorbid with a wide range of Axis I conditions. For example, a study of more than 18,000 patients seeking evaluation at a psychiatric clinic showed that almost 80% of those with a personality disorder received an Axis I diagnosis also (Fabrega et al., 1991). Disorders that overlap frequently with personality disorder include major

P.185

depression and dysthymia, panic and other anxiety disorders, somatoform disorders, substance abuse disorders, and eating disorders—in other words, almost every major class of disorder (Clark, Watson,& Reynolds, 1995). Moreover, the presence of personality disorder in these disorders is associated with greater severity of psychopathology, greater social impariment including poorer social support, and slower and worse response to treatment (Pfohl et al., 1991; Ruegg & Frances, 1995). As I discuss later, the degree to which the relations between personality disorder and temperament are intertwined with—versus independent of—this overlap with Axis I pathology remains unknown. Indeed, the precise degree of overlap and, more important, the meaning of the overlap among these disorders is quite controversial.

Mood in Personality Disorder

There are several ways in which mood is an important element in personality disorder. First, as mentioned earlier, emotionality or affectivity is considered part of the defining nature of traits. Second, for several of the personality disorders, the primary defining characteristic of the disorder itself is emotion-based. For example, the defining characteristic of borderline personality disorder is “a pervasive pattern of instability of interpersonal relationships, self-image and affects, and marked impulsivity” (APA, 1994, p. 654, emphasis added). In the specific criteria this is elaborated in two ways: “affective instability due to a marked reactivity of mood (e.g., intense episodic dysphoria, irritability, or anxiety usually lasting a few hours and only rarely more than a few days),” and “inappropriate, intense anger or difficulty controlling anger (e.g., frequent displays of temper, constant anger, recurrent physical fights)” (APA, 1994, p. 654). Similarly, histrionic personality disorder is defined as “a pervasive pattern of excessive emotionality and attention seeking” (APA, 1994, p. 657), which is then further specified as manifested in “rapidly shifting and shallow expression of emotions” and “self-dramatization, theatricality, and exaggerated expression of emotion” (APA, 1994, p. 658).

A third way in which mood plays a role in personality disorder is that almost all of the Axis II disorders contain at least one specific criteria that is affect-re-lated. table 9-4 lists the various personality disorder criteria that refer to affective phenomena, which include the specific affects or anger/irritability and anxiety in a range of disorders, as well as restricted emotional expression in schizoid and schizotypal personality disorders, affective deficits in antisocial and narcissistic personality disorders, and a variety of emotion-related criteria in several other disorders. It may be noted that the two personality disorders placed in the DSM-IV appendix for further study—depressive and negativistic personality disorders—also each have a number of affect-relevant criteria and, in the case of depressive personality disorder, affective disturbance is the primary feature of the disorder.

It may be apparent from these illustrations that the treatment of emotion in personality disorder is not simple, clear, or systematic. This disorganized state is not unique to emotion in the definition and assessment of personality disorder. There are fundamental problems in the diagnosis of personality disorder and the treatment of emotions in personality disorder is no exception. A basic problem is that personality disorders are defined categorically—as all-or-none types—whereas the traits of which they are composed are continuous dimensions

P.186

with no clear points of discontinuity to differentiate between disorders or even to demarcate normality from disorder. This disjunction of structure between on the one hand, distinct personality disorders and, on the other hand, then: continuous defining features, creates a tension in this still-evolving field of study. Increasingly, researchers are working to resolve this tension by integrating