The p Factor

An under-studied hypothesis about mental disorders

Mark your calendars: This Monday, May 1, Tim B. Lee of Full Stack Economics will join our Zoom for paid subscribers at 8 PM New York Time to talk about his new Substack, Understanding AI. Then on May 8 Bryan Caplan will join us to talk about his latest collection of essays, Voters as Mad Scientists.

In 2014, Avshalom Caspi and others published a paper arguing that the correlations among psychological disorders suggest the existence of a common factor. They proposed calling this the p factor. It is analogous to the g factor, known as general intelligence, popularly known as IQ.

I am surprised to find very little in the way of follow-up literature. When a striking and controversial theory appears in economics (take the Efficient Market Hypothesis, for example), the follow-on literature, pro and con, is voluminous.

I have not taken so much as one course in personality psychology. But the p-factor theory strikes me as interesting, in the same way that the EMH is interesting.

For one thing, the p-factor theory puts a focus on innate factors in determining mental health. But it suggests that parents with one disorder might have children with very different disorders.

Someone with a high p factor is likely to have parents with high p factors and/or experienced brain development that resulted in a high p factor. See Kevin Mitchell’s Innate.

For example, when I see an essay that refers to an old paper on mothers of boys with gender identity disorder showing that the mothers have dramatically higher incidence of borderline personality, I do not presume that the mothers’ behavior toward their sons caused their sexual identity issues. Instead, I think of the mothers as having a high p factor, which they pass along to their sons, and this happens to manifest itself in such issues.

When Jonathan Haidt wants to blame an increase in teen depression on an environmental factor (smart phones and social media), how does this relate to the p factor? One possibility is that the p factor predicts a propensity for mental disorders, but the environment heavily influences which mental disorders become manifest. Smart phones and social media do not affect the p factor, but they trigger an epidemic of teen depression, because that is how a high p factor manifests itself in the current environment. Before smart phones, the same people would have been susceptible to mental illness, but it might have manifested differently or not at all.

On a somewhat related note, Scott Alexander wrote,

Watters suggests, there probably is some base-level objectively-real mental illness. If you have to think of it as something, you can think of it as formless extreme stress, looking for an outlet. But the particular way the stress finds an outlet is based on the patient's cultural preconceptions.

He refers to Ethan Watters’ Crazy Like Us,

The p-factor theory can account for what I see as the difficulty in distinguishing among mental disorders. I have heard of the same child being diagnosed by one psychiatrist as autistic, by another as having defiance disorder, by another as having anxiety, and by another as having ADHD. This seems to me consistent with the idea that there is a general propensity for mental disorders that is manifested in different ways under different circumstances.

I should note that all of this is complicated by the fact that the term “disorder” is loaded. It has a connotation that there is something is wrong with the person. Making that judgment often feels uncomfortable. But I don’t think that we should completely avoid making such judgments.

Another possibility that occurs to me is that instead of (or in addition to) thinking in terms of a p factor that causes mental illness we might think of a factor that protects against mental illness making someone dysfunctional. This might be termed a “coping factor,” which we could call c. That is, if you have two people with a propensity for a given disorder, the one with higher c will be able to overcome the disorder, and the one with lower c will be unable to do so. So we have the high-functioning autistic compared to the low-functioning autistic. We have the creative and original thinker compared to the schizophrenic.

I wonder if some drugs that tend to work on mental disorders do so by raising c in some people. Note that the mechanisms by which anti-depression medications work have recently been called into question. The role of serotonin, in particular, has been cast into doubt. I wonder if perhaps the medications do not fight depression per se, but they give some people a higher level of c, which changes the way that people experience depressive episodes.

The p factor theory says that a single patient might well be subject to a variety of disorders. The c factor theory speaks to the mildness or severity of the disorder that will be manifest in the patient.

These sorts of speculation on my part are not well grounded. But if I were a student in the psych field, I would expect to find considerable research into the p factor theory (or even the c factor theory). Relative to what I would expect to find, the literature seems quite sparse.

This essay is part of a series on human interdependence.