Matthew Broome

The two-factor theory (Davies et al. 2001; Coltheart 2007; Coltheart, Menzies & Sutton 2010) is an influential account of delusion formation. According to the theory, there are two distinct factors that are causally responsible for delusion formation. The first factor is supposed to explain the content of the delusion, while the second factor is supposed to explain why the delusion is adopted and maintained. Recently, another remarkable account of delusion formation has been proposed, in which the notion of “prediction error” plays the central role (Fletcher & Frith 2009; Corlett et al. 2009; Corlett et al. 2010). According to this account, the prediction-error theory, delusions are formed in response to aberrant prediction-error signals, those signals that indicate a mismatch between expectation and actual experience .
Is the prediction-error theory a rival to the two-factor theory? Prediction-error theorists tend to be critical about the two-factor theory and present their views as an alternative to it. Fletcher and Frith wrote, on the two-factor theory: “symptoms reflecting false perception and false beliefs are so intertwined in schizophrenia that a theory relying on coincidental damage seems very unlikely.” (Fletcher & Frith 2009, 51) Again, Corlett and colleagues argue that positing two factors is redundant because “a single deficit in Bayesian inference is able to explain more of what we know about the interactions between perception and belief-based expectation, the neurobiology of the delusions that occur in schizophrenia and the maintenance of delusions in the face of contradictory evidence.” (Corlett et al. 2010, 357)
In this chapter, we examine the relationship between the two-factor theory and the prediction-error theory in some detail. Our view is that the prediction-error theory does not have to be understood as a rival to the two-factor theory. We do not deny that there are some important differences between them. However, those differences are not as significant as they have been presented in the literature. Moreover, the core ideas of the prediction-error theory may be incorporated into the two-factor framework. For instance, the aberrant prediction-error signal that is posited by prediction-error theorists can be (or underlie) the first factor contributing to the formation of some delusions, and help explain the content of those delusions. Alternatively, the aberrant prediction-error signal can be (or underlie) the second factor, and help explain why the delusion is adopted and maintained.

We offer two other examples of the way in which neurosciences, at the current stage of development, cannot do all the explanatory work in psychiatry. The first example centres on the employment of normative notions in the characterization of the manifestation of psychopathologies as deviant. The second example highlights the importance of environmental factors in the onset of psychosis. In the end, we explore some potential for future research in these areas.