The Two-Factor Theory of Delusions

What it is

The division of labour is what makes the theory attractive. Factor one explains the content of the delusion. Factor two explains why it is held. Without the second factor, an odd experience produces puzzlement rather than a fixed false belief, and most people who have odd experiences do in fact end up puzzled.

In Capgras delusion, the standard example of factor one is the loss of the normal autonomic response to a familiar face, so that the face is recognised and does not feel familiar. The hypothesis that the person is an impostor accounts for the experience. What has to be explained separately is why it is not then rejected.

The originating paper is Davies, Coltheart, Langdon and Breen, "Monothematic Delusions: Towards a Two-Factor Account", in Philosophy, Psychiatry and Psychology in 2001. Coltheart developed it in his 2007 Bartlett Lecture and in Coltheart, Langdon and McKay in the Annual Review of Psychology in 2011.

The weakest leg is the localisation. The literature places factor two in the right dorsolateral prefrontal cortex, inferred largely from the observation that monothematic delusions after brain injury tend to follow right-sided damage. Coltheart has himself said the localisation is a hypothesis rather than an established finding, and it was tested directly in Coltheart, Cox and colleagues in Cortex in 2018, using transcranial magnetic stimulation, where the result did not support a straightforward causal role. An originator running the experiment that undercuts his own localisation is a standard worth quoting at other people.

In effect

The other side deserves stating properly. One-factor theorists argue that a single anomalous experience plus normal reasoning is sufficient, and that a second factor is an unnecessary posit. Prediction-error and predictive-coding accounts offer a rival framework that does not require a discrete belief-evaluation module at all, treating delusion instead as an aberrant weighting of prediction error. The argument over what is even meant by one factor against two is live in cognitive neuropsychiatry.

There is also a correction this page exists to carry. Marc Dingman's Bizarre uses the second factor as a load-bearing device across three chapters and names nobody. He calls it a plausibility-checking mechanism, introduces it in chapters 1 and 2, and applies it through chapter 9 to Capgras, Cotard, Fregoli and mirrored-self misidentification. That is the two-factor account exactly, applied to the same syndrome family, with no name and no primary paper attached.

The right thing to say about that is precise. To his credit he labels the device as hypothetical at pages 17 and 18; from page 22 onward he uses it as settled machinery, which is a different thing. The body text does not credit the originators. Whether the endnotes do could not be established in this publication's verification pass. It is recorded as not credited in body text, with endnote status unverified, and never as anything more than that.

What it does not say

It is not settled. It is the most developed framework, with one-factor and predictive-coding accounts as live rivals.

It does not have an established neural location. The right dorsolateral prefrontal localisation is a hypothesis, and the direct test of it did not support a straightforward causal role.

It does not rest on a strong evidence base. The syndromes it was built on are documented largely in case reports.

It does not belong to whoever last used it. The account is Davies, Coltheart, Langdon and Breen's, and this page credits first authors accordingly.

---