The General Factor of Psychopathology and Personality

Snyder, Young, and Hankin (this issue) provide further support for a general factor of psychopathology. However, they acknowledge, “more importantly, it will be important for future research to continue to understand the nature of the p factor.”

We suggest a comparison with the general factor of personality. A predominant personality trait model is the Big Five, consisting of negative emotionality, extraversion (versus introversion), openness (or unconventionality), agreeableness (versus antagonism), and conscientiousness (versus disinhibition). Above the Big Five are the Big Four, Three, Two, and even a Big One.

How best to understand the first general factor of personality though is in dispute. At the level of the Big Five, traits align in a conceptual manner. Traits that are opposite in meaning anchor opposing poles (e.g., introversion vs. extraversion). However, this conceptual arrangement is severely disrupted at the level of the Big One.

One understanding of the Big One is that it is artifactual, reflecting simply a disposition to evaluate oneself in a positive or negative manner (Pettersson, Turkheimer, Horn, & Menatti, 2012). The predominant rationale for this understanding is that traits that are essentially opposite to one another (e.g., unambitious and workaholic) load comparably. Persons can not be endorsing the presence of opposite traits if they are providing accurate self-description. Proponents of this view suggest that variance due to evaluation should be extracted from the lower order factors in order to obtain more unbiased and substantively interpretable scales.

The evaluation bias explanation is a rebirth of the social desirability hypothesis promulgated years ago. Scales were even developed to assess for this impression management. This hypothesis though was ultimately discredited (McCrae & Costa, 1983). It is evident that most persons are providing reasonably accurate and honest self-descriptions. It would be quite unlikely that such a large degree of variance would reflect simply impression management. Individual differences on social desirability scales reflect true individual differences in positive versus negative attributes.

Indeed, the Big One is probably a reflection of adaptivity versus maladaptivity, as this is the primary, if not the only, meaningful way to align all traits along a single common dimension. One can not conceptually align traits that have no shared meaning along a common latent dimension (e.g., traits from different FFM domains are unrelated conceptually to one another). However, one can align these traits with respect to their implications for adaptive functioning.

Some suggest that the Big One reflects a true substantive dimension, reflecting (for instance) a degree of survival fitness or general emotional satisfaction (Musek, 2007). This is comparable to the proposals within psychopathology research for a nonspecific psychopathological disposition that subsequently becomes more differentiated into more specific forms through the course of development (Caspi et al., 2014).

There has been some interest in developing a nomenclature of psychopathology in which disorder is diagnosed independently of dysfunction. There is also a comparable interest in personality, to assess traits independent of implications for impairment. To the extent that this could be done, it might indeed be the case that the etiology for the traits (i.e., their location within the Big Five) is different from the etiology for their dysfunction (i.e., their location within the Big One).

However, assessing for the presence of traits (or disorder) separate from dysfunction is proving to be exceedingly difficult, if not impossible. One can not assess (for instance) irresponsibility or gullibility without implicitly, if not explicitly, assessing for their maladaptive consequences. A few traits can be assessed in a neutral manner, but these appear to be the exceptions. Dysfunctional (or functional) outcome could very well be inherent to the presence of particular traits (and disorders).

If traits and dysfunction can not be separated, one is again left with the difficulty of explaining the Big One. We suggest that the alignment within the Big One simply may reflect, at least in part, a nonspecific impairment or dysfunction, irrespective of the etiology or basis for this impairment.

It is not difficult to understand that adaptive functioning will typically correlate positively with other indicators of adaptive functioning, irrespective of the content, bases, or etiology for this dysfunction; likewise, one form of dysfunction will correlate positively with other forms of dysfunction. Anyone who has constructed a nonspecific, generic functioning scale will be familiar with this general point.

This might be part of, if not the essential, basis for the Big One in psychopathology and personality. The traits (and disorders) have to be aligned in some manner within the general factor. They can not be aligned with respect to their conceptual meaning or content. The only choice is their implications for dysfunction.

The alignment is then artifactual (i.e., independent of the traits or disorder) albeit not in the same meaning or sense of the evaluation hypothesis. In addition, although artifactual this alignment is not weak or trivial. On the contrary, extent of dysfunction and impairment is a very powerful basis for aligning (correlating) variables, constructs, and measures. First extracted factors will at times concern simply impairment because dysfunction can account for much of the correlation amongst a battery of scales.

Indeed, the nature or extent of dysfunction can work against the conceptual meaning of a respective trait. Traits that are opposite to one another (e.g., lax and perfectionistic) can be positively correlated because they result in the same degree and perhaps even the same nature of negative outcomes (e.g., poor work performance). One should not though control for this problem by partialing out the variance due to the Big One, because the maladaptivity is inherent to the traits. Removing this variance will alter the fundamental nature of the traits (or disorder) in such a manner that their original meaning or understanding is distorted, a lesson learned many years ago with respect to social desirability.