The Neuropsychological Model of Temperament Developed by Gray

Panic Away

Panic Away End Anxiety and Panic Attacks

Get Instant Access

In the 1970s Eysenck's most prominent student, Gray, undertook a thorough critique of his master's theory of extraversion and neuroticism, which led to the development of Gray's own neuropsychological theory of temperament or personality (Gray used the terms interchangeably; Gray, 1991). This theory differs from all other temperament conceptualizations in that it is the most physiologically oriented one, supported by rich neurophysiological, pharmacological, and biochemical evidence, collected, however, mainly in studies on rats.

Applying the four criteria by which temperament theories have been classified (see Table 3.1), Gray's neuropsychological model oftemperament is seen as oriented to adults, and a causal, multidimensional; and emotion-centered theory. His temperament model has developed over the past two decades and has been described in many papers and books. To write about the biological basis of personality or temperament without reference to Gray's contribution reflects ignorance.

Theoretical Background

From the beginning ofhis research career Jeffrey Gray was involved in studying arousal as related to temperament. In his Ph.D. thesis, written in 1968 under the supervision of Peter Broadhurst (in Eysenck's laboratory), he studied the relationship between level of arousal and behavioral indices of extraversion (cited in H. J. Eysenck & Eysenck, 1985). Eysenck's initial interpretation of extraversion in terms of Pavlov's concepts of excitation and inhibition motivated Gray (1964b) to undertake a thorough examination of the Pavlovian and neo-Pavlovian typology, with special reference to the construct of strength of excitation. In order to bring the concept of strength of excitation to the attention of biologically oriented personality researchers, Gray (1964c) undertook a reinterpretation of strength of excitation in terms of arousal and arousability; the latter construct was introduced by him (for details see Chapter 5). His theoretical considerations regarding arousal were based on a profound knowledge of research in this domain, with special reference to the contributions of Duffy, Malmo, Hebb, Berlyne, and Lindsley (for references see Chapter 5). Gray was also deeply involved in studying classical and operant conditioning, particularly as related to behavior and behavior disorders. His own research on rats (Gray, 1967; Gray & Smith, 1969) showed the role ofre-ward and punishment in animal learning and made him aware of the importance of individual sensitivity to the effects of reward and punishment in classical and operant conditioning (Gray, 1975). Gray's studies on animal learning were deeply rooted in the learning theories of Pavlov (1951-1952), Hull (1952), Mowrer (1 960), Amsel (l962), and Spence's (Spence & Spence, 1966) learning theory of anxiety, as well as on empirical evidence regarding the role of brain structures as rewarding or punishing loci (McCleary, 1966; Olds & Olds, 1965).

Although he concentrated on experimental research concerning the physiological and biochemical mechanisms and neural structures underlying anxiety in rats, Gray used animal data to develop a neuropsychological theory of human temperament. His studies on temperament were also strongly influenced by Eysenck and the Russian neo-Pavlovian typologists Teplov and Nebylitsyn.

Anxiety and Impulsivity: The Two Basic Temperament Dimensions

Gray (1970, 1981) undertook a major revision of Eysenck's theory of temperament, mainly by questioning the conditionability paradigm and the physiological mechanisms mediating extraversion and neuroticism (see Chapter 2). Analysis of data collected in the domain of eyeblink conditioning in man as related to extraversion and neuroticism, and studies on rats which showed that barbiturate drugs and alcohol, similarly to lesions of the frontal cortex, tend to lead to extraverted behavior, led Gray to the following conclusions: (a) Introverts are more susceptible to punishment and nonrewards than are extraverts, whereas extraverts are more sensitive to rewards and nonpunishment; and (b) neuroticism is a dimension which may be characterized in terms of general sensitivity to reinforcing events, both rewards and punishments; neurotics, in comparison with emotionally stable individuals, are characterized by increased sensitivity to rewards and punishments.

The critical approach to extraversion and neuroticism, and studies on the biological mechanisms of anxiety in rats, including his own findings, allowed Gray to conclude that extraversion and neuroticism are secondary traits emerging from a combination (interaction) of two basic temperament traits—anxiety and impulsivity—as illustrated in Figure 3.4. These two traits, orthogonally related, are a result of a 45o rotation of Eysenck's neuroticism and extraversion.

Susceptibility to reward Susceptibility to punishment

FIGURE 3.4. Relation between sensitivity to stimuli associated with reward (impulsivity) and with punishment (anxiety), neuroticism and introversion-extraversion. Note. From "The Neuropsychology ofTemperament," by J. A. Gray. In J. Strelau and A. Angleitner (Eds.), Explorations in Temperament: International Perspectives on Theory and Measurement (1991, p. 124). Copyright 1991 by Plenum Press. Reprinted with permission.

Susceptibility to reward Susceptibility to punishment

FIGURE 3.4. Relation between sensitivity to stimuli associated with reward (impulsivity) and with punishment (anxiety), neuroticism and introversion-extraversion. Note. From "The Neuropsychology ofTemperament," by J. A. Gray. In J. Strelau and A. Angleitner (Eds.), Explorations in Temperament: International Perspectives on Theory and Measurement (1991, p. 124). Copyright 1991 by Plenum Press. Reprinted with permission.

Influenced by N. E. Miller's (1951), Amsel's (1962), and Mowrer's (1960) animal learning theories which postulate that emotions consist of states elicited by stimuli having the capacity to act as reinforcers for instrumental behavior, Gray developed two assumptions which guided his studies on temperament, and postulated

(1) that temperament reflects individual differences in predispositions towards particular kinds of emotions and (2) that emotions are states of the CNS elicited by reinforcing events. . . .Reinforcer is any stimulus (or more complex event) which, if made contingent upon a response, alters the future probability of emission ofthat response. (Gray, 1991, pp. 106-107)

Anxiety as a trait is mediated by sensitivity to signals of punishment, nonreward, and novelty. High sensitivity determines a high level of anxiety (see Figure 3.4). Individual differences in impulsivity depend on sensitivity to signals of reward and nonpunishment. The higher the sensitivity to these signals, the higher the impul-sivity (Gray, 1981).

Taking this model as a point of departure, extraversion may be characterized as being composed of low anxiety and high impulsivity, whereas neuroticism is a combination of high anxiety and high impulsivity (see Figure 3.4). In other words, extraverts are characterized by high sensitivity to signals of rewards and low sensitivity to signals ofpunishment, whereas introverts are more sensitive to punishment than to reward. As regards neurotics, effects of all reinforcing stimuli (punishment and reward) are amplified. Figure 3.4 illustrates also that neurotics differ from emotionally stable individuals in that they are more susceptible to both kinds of conditioned stimuli (punishment and reward) than are emotionally stable individuals.

In his recent publications, Gray (1991) postulated further the existence of a third basic temperament trait which may be called defensiveness. This trait is expressed in defensive aggression or escape bahavior. As opposed to anxiety and impulsivity, which are mediated by sensitivity to conditioned reinforcers, defen-siveness is determined by susceptibility to unconditioned aversive stimuli.

Susceptibility of the BIS, BAS, and F/FLS as Biological Bases ofTemperament Traits

Gray's most significant contribution to research on temperament consists in his development of empirically grounded hypotheses regarding the biological mechanisms mediating temperament characteristics. As mentioned in the preceding citation, emotions, to which temperament refers, are states of the CNS. By the abbreviation CNS Gray (1987, 1991) intended to underline that in his physiological studies on temperament he is referring to the "central nervous system" as well as to the "conceptual nervous system," for both of which this abbreviation holds.

Gray hypothesized that in the brain there are three distinguishable systems for the control of emotional behavior, composed of specific structures and functions that constitute the neurological basis for the three temperament dimensions, each of these dimensions "corresponding to individual differences in the sensitivity or reactivity of one emotion system" (Gray, 1987, p. 494). I limit the present account to more general issues concerning the physiological mechanisms underlying the three basic emotions and the corresponding emotion-traits. The details which refer to neurophysiological, biochemical, and pharmacological data may be found in Gray's (1982a, 1982b, 1991; Gray, Owen, Davis, & Tsaltas, 1983; Gray, Feldon, Rawlins, Hemsley, & Smith, 1991) publications.

The Behavioral Inhibition System. Sensitivity to signals of punishment, non-reward, and novel stimuli, as well as reactions to these stimuli (behavioral inhibition, increment in arousal, and increased attention) regarded as expressions of anxiety, are mediated by the Behavioral Inhibition System (BIS). The core structure for BIS is the septohippocampal system which comprises the following three basic neural structures: the hippocampal formation, the septal area, and the Papez circuit.1 BIS activity is accompanied by a subjective state identified as anxiety.

1Some details regarding the anatomy, physiology, and biochemistry of the central nervous system are presented in Chapter 4.

This is "a state in which one responds to threat (stimuli associated with punishment or nonreward) or uncertainty (novelty) with the reaction, 'stop, look, and listen, and get ready for action"' (Gray, 1991, p. 110). Reactivity (susceptibility) of the operating parameters of BIS (e.g., thresholds, ease of excitation, speed of operation) determines individual differences in anxiety as a trait (Gray, 1983).

BIS should be regarded as a unified system that may be identified by means of pharmacological treatment. The administration of drugs, such as alcohol, barbiturates, and benzodiazepines, on the one hand, causes a decrease in the susceptibility of the BIS and, on the other, reduces anxiety (Gray, 1982a, 1982b). In turn, ascending monoaminergic pathways (both noradrenergic and serotonergic) originating in the brain stem have a boosting effect on the activity of the septohip-pocampal system. These pathways are activated in stress-inducing situations, especially those which are of biological significance (Gray, 1983).

The Behavioral Approach (Activation) System. Much less elaborated are the physiological and neuroanatomical bases for impulsivity which Gray identified with reactivity of the Behavioral Approach System (BAS), often termed also the Behavioral Activation System (Fowles, 1980). Recently Gray (1991), in discussing the temperament dimension corresponding to BAS, hypothesized that high sensitivity (reactivity) of BAS may be associated with behavior motivated by positive reinforcement, thus related to positive emotionality which comprises such emotions as hope, happiness, and elation.

The neurological structure of BAS consists of the following hypothesized key components: basal ganglia, dopaminergic fibers ascending from the mesencephalon, thalamic nuclei, and neocortical areas (motor, sensorimotor, and prefrontal cortex), all ofthem interacting with each other. They constitute two motor systems: the caudate motor system and the accumbens motor system. The function of the caudate motor system is to encode the relationships between input (signals of reward and nonpunishment) and output (responses) in a goal-directed motor program. The accumbens motor system is responsible for switching between the consecutive steps in the motor program (Gray, 1991; Gray et al., 1991). The motor program is guided by the projection to nucleus accumbens from the amygdala. The BAS closely interacts with the BIS in that the septohippocampal system controls the matches between the actual outcome and expected outcome of a particular motor step.

Release of dopamine from the terminals of neurons whose cell bodies are in the ventral tegmental area, and especially in the nucleus accumbens, activates the BAS. As a consequence, dopamine release elicits approach behavior similar to that which occurs as a response to positive reinforcers such as food, a sexually receptive partner, and chemically diverse drugs such as amphetamines, nicotine, alcohol, heroin, and cocaine (Gray, 1991; Gray et al., 1991).

The Fight/Flight System. As already mentioned, Gray (1987, 1991) in his recent publications distinguished a third system for the control of emotional behav ior: the Fight/Flight System (F/FLS). Reactivity of F/FLS determines individual differences in defensiveness, a basic temperament trait that may be identified in human beings as anger.

F/FLS responds to unconditioned aversive stimuli with unconditioned defensive aggression or escape behavior. There are three major neural structures of which F/FLS is composed: amygdala, medial hypothalamus, and central gray. The septohippocampal system, through its connections with the medial hypothalamus, influences the functioning of F/FLS. The medial hypothalamus inhibits the final output pathway in the central gray; the amygdala inhibits the medial hypothalamus (Gray, 1991). Some authors (e.g. Adams, 1979; Panksepp, 1982) made detailed analyses of the anatomy involved in the F/FLS.

As can be seen from this short presentation of the neuropsychological basis of temperament dimensions, "much of the above analysis is inevitably speculative" (Gray, 1991, p. 113). The hypothetical constructs of BIS, BAS, and F/FLS, although referring to structures and functions of the central nervous system, are themselves conceptual nervous system concepts.

Psychometric Attempts to Measure Temperament Traits Elicited on the Basis of BIS, BAS, and F/FLS Constructs

Starting from the assumption that the three neuropsychological systems responsible for the control of emotional behavior make it possible to predict the structure of temperament, G. D. Wilson, Barrett, and Gray (1989) constructed an inventory directly related to the inputs and outputs of BIS, BAS and F/FLS. This inventory, known as the Gray-Wilson Personality Questionnaire (GWPQ), consists of the following six scales: (1) Approach and (2) Active Avoidance which are deduced from the BAS construct, (3) Passive Avoidance and (4) Extinction which are supposed to refer to BIS, (5) Flight and (6) Fight which are related to F/FLS. The GWPQ is composed of 120 well-balanced items, 20 for each scale. A study conducted on 243 adult men and women has given reliability scores varying from .35 (Active Avoidance in women) to .71 (Approach in men, and Fight and Flight in women), which should be considered unsatisfactory. Most important, however, is the fact that the intercorrelations of the six scales did not correspond with the predictions of Gray's theory. For example, approach and active avoidance should correlate positively, but the results were just the opposite. Also, Fight and Flight, having the same neurological basis, were expected to correlate positively but the results obtained in this study did not confirm this assumption. From their data the authors concluded that "these results present a difficulty for Gray's theory of personality as applied to human subjects" (Wilson et al., 1989, p. 513). Sosnowski and Bialski (1992) also failed in a questionnaire investigation of Gray's three-dimensional theory. A correlational and factor analysis of items from the six scales of which their inventory was constructed supported the existence of two or thogonal factors, similar to impulsivity and anxiety. S. A. Ball and Zuckerman (1990) were more successful in developing a two-factor scale based on Gray's theory. Their Generalized Reward and Punishment Expectancy scales (GRAPES) showed reasonable convergent validity when related to the SSS and EPQ-R scales.

Final Remarks

Except for the unsatisfactory GWPQ there is no reasonable diagnostic measure that enables an assessment to be made of the temperament traits that directly correspond with the three physiological systems controlling emotional behavior, as postulated by Gray. Among the most popular diagnostic instruments aimed at measuring the BIS and BAS constructs are EPQ-R, the State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, & Lushen, 1970) and Cloninger's Tridimen-tional Personality Questionnaire (TPQ; Cloninger, Svrakic, & Przybeck, 1991). This has been demonstrated in several publications (see, e.g. Boddy, Carver, & Rowley, 1986; Derryberry, 1987; Gray et al., 1991; MacAndrew & Steele, 1991).

Pickering (1997) reviewed his own studies in which the aforementioned inventories as well as some other psychometric instruments suggesting associations with BIS and BAS were applied. He arrived at a conclusion that the findings regarding links between BIS and BAS and other temperament dimensions which are supposed to be related to these constructs are contradictory.

Comparing Gray's theory with his own, Eysenck concluded that, in fact, they are very similar. One of the main differences lies in their different predictions. Whereas Eysenck's theory predicts superior conditioning in introverts when appetitive (rewarding) unconditioned stimuli are used, Gray expects extraverts to be more susceptible to rewards, thus more efficient in conditioning (H. J. Eysenck & Eysenck, 1985). These differences derive from different physiological interpretations postulated by the two authors. A vigorous critique of Gray's theory from an Eysenckian point of view was made by Robinson (1986) who came to the extreme conclusion that "Gray's theory is really not much more than a theory of anxiety. So much so that it can barely claim to be a theory of personality at all; let alone one that might replace the Eysenckian paradigm" (p. 467).

If we look at Gray's neuropsychological model of temperament from the standpoint of human temperament, we need to bear in mind that this model was developed on the basis of animal studies, centered mainly on the physiological bases of rat anxiety (Gray, 1978, 1982a, 1982b). The failure to develop a psychometric measure of human temperamental traits based on Gray's animal learning paradigms exemplifies the difficulties and problems in making human-animal analogies or comparisons in this domain of study (see Gray, 1973; Robinson, 1986). This does not deny the fact that Gray's extensive studies on the biological basis of temperament in animals are of great importance for an understanding of the neurophysiological backgrounds of human temperament (see, e.g., Cloninger, 1986; H. J. Eysenck & Eysenck, 1985; Rothbart, 1989a; Zuckerman, 1991c.)

Zuckerman (1991c), in his biological interpretation of a variety of biologically based personality dimensions, has most convincingly shown how useful Gray's neuropsychological model of temperament is for bringing us closer to an understanding ofthe nature of temperament. Zuckerman's main criticism of Gray's theory was also directed toward the rat bias in his approach to temperament, in which behavior was limited mainly to habituation, conditioning, and reinforced instrumental learning. The spectrum of the human personality expressed in social behavior is different from animal behavior in its variety and content. To take physiological mechanisms as a starting point in studying temperament dimensions leads to a "bottom-up" approach. This approach, which Gray represents, is based on the assumption that temperament dimensions and specific neurological and neurotransmitter systems are isomorphic (Zuckerman, 1991c, 1992). There is, however, no empirical evidence for temperament-biology isomorphism.

Several authors aim their criticism at the biological model underlying temperament traits. A detailed critique of this model was presented in an Open Peer Commentary to Gray's 1982b paper. This critique shows that experts who study the anatomical, physiological, and biochemical basis of anxiety, although recognizing Gray's major contribution to this field of study, are far from agreement (expressed even in contradictory statements) regarding the answer to the question: "What are the neuropsychological bases of anxiety?"

Gray's neuropsychological approach to temperament, which to my knowledge is the approach most deeply rooted in experimental data and in the neurological sciences and biochemistry, exemplifies, better than any other temperament theory, how far we still are from answering the question posed in the preceding paragraph.

Was this article helpful?

0 0
Kicking Fear And Anxiety To The Curb

Kicking Fear And Anxiety To The Curb

Kicking Fear And Anxiety To The Curb Can Have Amazing Benefits For Your Life And Success. Learn About Calming Down And Gain Power By Learning Ways To Become Peaceful And Create Amazing Results.

Get My Free Ebook


Post a comment