THE PEN MODEL

Descriptive Aspects

The PEN model, proposed and advocated by Eysenck as the overarching paradigm of personality psychology, has two main aspects: descriptive, and causal. The descriptive aspect of the model is a hierarchical taxonomy based on factor analysis. At the top of the hierarchy are the superfactors of psychoticism, extraversion, and neuroticism (PEN). These superfactors are comprised of factor analyses of lower-order factors such as sociability and positive affect (components of extraversion). These factors are comprised of factor analyses of low-order habits such as liking to study with a group of people (a component of sociability). These habits are comprised of factor analyses of lower-order behaviors such as studying for the personality midterm with a group of people.

Two important principles of personality research that are incorporated into the PEN model are aggregation and the state-trait distinction. The principle of aggregation is that measures will have higher reliability if they are comprised of many items. For example, extraversion is comprised of many different factors, habits, and behaviors, and therefore should have good reliability. The state-trait distinction is also built into the PEN model. At the top level, the superfactors of P, E, and N are traits that are very stable across time and situation. At the bottom level, the behavior of studying for the midterm with a group of people is a state that could vary widely, for example, with the availability of study partners. While states are very changeable, traits are very stable.

Eysenck believes that the five-factor model is a hodge-podge of factors and superfactors. The five-factor model and the PEN model both include extraversion and neuroticism at the highest level. However, the superfactor of psychoticism is made up of the lower-level factors of (dis-) agreeableness and (non-) conscientiousness. Moreover, the big five include "intellect" (self- or peer rated) at the top level. Eysenck thinks of this as simply a bad way of measuring intelligence, which is a cognitive ability that would be better reflected in I.Q. tests than in self-reports on adjective questionnaires.

One factor that was originally subsumed under the superfactor of extraversion in the Eysenck Personality Inventory (EPI) is impulsivity. Later, when the dimension of psychoticism was added, impulsivity was moved from extraversion to psychoticism in the Eysenck Personality Questionnaire (EPQ). Some investigators, such as Gray and Revelle, believe impulsivity is uniquely important, and that its removal from the extraversion superfactor is a bad idea.

Causal Aspects

Extraversion. The PEN model is biologically based. Extraversion is based on cortical arousal. Arousal can be measured by skin conductance, brain waves, or sweating. While theoretically introverts are chronically overaroused and jittery, theoretically extraverts are chronically underaroused and bored. The theory presupposes that there is an optimal level of arousal, and that performance deteriorates as one becomes more or less aroused than this optimal level. The finding that arousal is related to performance as an inverted U-shaped curve is called the Yerkes-Dodson Law.

Extraversion is related to social interest and positive affect. Some investigators have proposed that social interest causes positive affect, since the best of times are usually those spent with other people. However, Diener and Larsen (1993) have found that this hypothesis is incorrect. Another alternative is that positive affect causes social interest, since being very enthusiastic and fun loving may make people want to go out and be with other people. This hypothesis has not yet been studied. Yet another possibility is that a third factor causes both positive affect and social interest. Dopamine responsivity, which makes people highly sensitive to reward, may be the factor responsible for both positive affect and social interest.

Neuroticism. Neuroticism is based on activation thresholds in the sympathetic nervous system or visceral brain. This is the part of the brain that is responsible for the fight-or-flight response in the face of danger. Activation can be measured by heart rate, blood pressure, cold hands, sweating, and muscular tension (especially in the forehead). Neurotic people, who have a low activation threshold, experience negative affect (fight-or-flight) in the face of very minor stressors--i.e., they are easily upset. Emotionally stable people, who have a high activation threshold, experience negative affect only in the face of very major stressors--i.e., they are calm under pressure.

It is interesting to note that measures of activation are not highly correlated. That is, people differ in which responses are influenced by stress--some sweat, others get headaches. This is called individual response specificity. It is also interesting to note that stressors differ in the responses they elicit. This is called stimulus response specificity.

Psychoticism. Psychoticism is associated not only with the liability to have a psychotic episode (or break with reality), but also with aggression. While less research has been done on psychoticism than on extraversion and neuroticism, the research that has been done has indicated that psychoticism too has a biological basis: increased testosterone levels.

In order to draw causal conclusions, researchers on the PEN model have not been content to use only correlational research methods such as factor analysis, but have gone further and used experimental research methods. These methods have been used not only on humans, but also on non-human animals such as rats. (Isn't it interesting that Eysenck believes even rats have personality--with the same three dimensions as humans?!)

Activity: Eysenck thinks the PEN model is a "paradigm," or a great idea in personality. What exactly is Eysenck's theory, and do you agree with his evaluation of it?

Back to Notes

Back to the Great Ideas in Personality Home Page