LB: Presentations of the two-factor account of delusion formation usually begin with two questions. The first question is about where the content of the delusion came from and the second is about the adoption or persistence of the belief. The two factors are supposed to provide answers to these two questions. But it sounds as if you are distinguishing a question about adoption from a question about persistence (or maintenance). If there are more than two questions to be answered, will an explanation of a delusion have to appeal to more than two factors?
MKD: Thank you for raising this issue of the relationship between questions and factors. It is quite important for understanding the two-factor framework. Questions about the aetiology of a delusion can be multiplied and, correspondingly, explanations of a delusion can be increasingly detailed, appealing to more than two ‘factors’ in the ordinary sense of that term. But, crucially, factors in the sense relevant to the two-factor framework are not just elements in an explanation. They are, specifically, pathologies or departures from normality, such as neuropsychological deficits.
You are right to say that I want to distinguish between a question about adoption of the delusional belief and a question about the belief’s persistence. If we are going to say something substantive about the nature of the second factor then we need to be clear about the role of the second factor. Where, in the total story of the aetiology of a delusion, does it figure? So, along with the first question about the source of the delusional idea or hypothesis, there should be the adoption question, ‘Why is the delusional hypothesis adopted as a belief?’, and the persistence question, ‘Why does that belief, once adopted, persist; why is it not subsequently rejected?’.
I think it is widely agreed that explaining a delusion requires answers to all three of these questions. But that does not settle the issue whether more than two factors are required in total. Perhaps one or other of the adoption and persistence questions can be answered without appeal to a departure from normality. For example, Max Coltheart, Peter Menzies and John Sutton (2010) argued, using Capgras delusion as their example, that no second departure from normality is required to answer the adoption question, but that a second factor is required to answer the persistence question.
Andy Egan and I (2013) argued for a similar view, at least for some cases of delusion. In response to Coltheart and colleagues (2010), Ryan McKay (2012) argued that a second factor is required to answer the adoption question, but that no third departure from normality is required to answer the persistence question. Another possibility is that both questions can be answered by appeal to the same departure from normality.
So it is certainly possible that three questions might be answered without appealing to more than two factors. But it is also possible that, in some or all cases of delusion, answering the adoption and persistence questions might require appeal to two different departures from normality – so that three factors would figure in the overall explanation of the delusion (consistent with the at-least-two-factor theory, but not with the exactly-two-factor theory).
MKD: Perhaps some recent work by Matthew Parrott (summarised here) suggests that there might be. We can think of the process of hypothesis evaluation as comparing the merits of a small number of competing explanatory hypotheses – a short list of live candidates. But then, it might be said, some hypotheses are so bizarre or so implausible that, normatively, they should not be considered as live candidates at all. For example, Parrott (in press) has proposed that an explanatory hypothesis is not worth considering seriously if it is already excluded by what one knows; that is, if it is ‘epistemically impossible’. So, alongside the adoption and persistence questions, the live candidate question arises: Why is the delusional hypothesis – even if it comes to the patient’s mind – not excluded at the very outset on the grounds that it is not a live candidate?
Even if the delusional hypothesis comes to the patient’s mind as a result of the anomalous experience (explained by the first factor), there are three distinct opportunities to reject it: it could be excluded from the short list at the outset; it could be rejected in a process of hypothesis evaluation on the grounds that there are stronger candidates on the short list; or it could be rejected in a process of post-adoption belief evaluation. In a case of delusion, how many departures from normality are required to explain this threefold failure of rejection? The two-factor theorist (specifically, the exactly-two-factor theorist) must answer, ‘Just one’.
The third part of this interview with Martin Davies will be published next Thursday.
Coltheart, M., Menzies, P. and Sutton, J. 2010: Abductive inference and delusional belief. Cognitive Neuropsychiatry, 15, 261–87.
Davies, M. and Egan, A. 2013: Delusion, cognitive approaches: Bayesian inference and compartmentalisation. In K.W.M. Fulford, M. Davies, R.G.T. Gipps, G. Graham, J. Sadler, G. Stanghellini and T. Thornton (eds), The Oxford Handbook of Philosophy and Psychiatry. Oxford: Oxford University Press, 689–727.
McKay, R. 2012: Delusional inference. Mind & Language, 27, 330–55.
Parrott, M. in press: Bayesian models, delusional beliefs, and epistemic possibilities. British Journal for the Philosophy of Science.