Two challenges that face popular self-monitoring theories (SMTs) of auditory verbal hallucination (AVH) are that they cannot account for the auditory phenomenology of AVHs and that they cannot account for their variety. In this paper I show that both challenges can be met by adopting a predictive processing framework (PPF), and by viewing AVHs as arising from abnormalities in predictive processing. I show how, within the PPF, both the auditory phenomenology of AVHs, and three subtypes of AVH, can be accounted for.
The positive symptoms of schizophrenia include delusions of control (“somebody else is controlling my actions”), thought insertion (“somebody is putting their thoughts into my head”) and auditory verbal hallucination (AVH) (hearing voices in the absence of a speaker). Perhaps the most popular theories for understanding these disparate symptoms are self-monitoring theories, which attempt to explain them as the product of one abnormality, namely, a problem with self-monitoring. According to these theories, our nervous systems distinguish self-generated from externally generated stimuli, through a process of self-monitoring. When this monitoring goes awry, self-generated stimuli are erroneously attributed to an external cause. The various positive symptoms all involve faulty monitoring and simply differ insofar as that which is failing to be properly monitored differs. In delusions of control it is bodily action, whereas in AVH and thought insertion it is widely thought to be inner speech (Feinberg, 1978, Frith, 1992, Jones and Fernyhough, 2007 and Seal et al., 2004). Although ingenious, the breadth of application of SMTs has recently been questioned by several theorists (Gallagher, 2004, Jones, 2010, Stephens and Graham, 2000 and Wu, 2012). Such criticisms tend not to take issue with the application of SMTs to symptoms involving bodily action, like delusions of control and (their merely experiential analogue) illusions of passivity. Rather, they claim that SMTs struggle to account for AVH and thought insertion. In this paper, I will focus on AVH, and on two challenges in particular. They are:
The Auditory Phenomenology Challenge – How do you explain the auditory phenomenology of AVH if it is misattributed inner speech?
The Varieties of AVH Challenge – How do you account for the varieties of AVH if it is (always) misattributed inner speech?
In this paper, I suggest that both challenges can be met if we adopt a recently popular general framework for thinking about what the brain does (e.g. Clark, 2013, Friston, 2005, Friston, 2010 and Hohwy, 2013) which we could call the predictive processing framework (PPF).
It is worth mentioning that the application of predictive processing to psychosis is not new. Indeed, Chris Frith, perhaps the best-known proponent of SMTs, has suggested something along these lines in Fletcher and Frith (2009). Since then, Adams, Shipp, and Friston (2013) have also suggested accounts of psychosis within the PPF. This work, however, does not focus on AVHs to the extent that I do, nor does it focus on the two challenges that I address here.
I proceed as follows. I start by presenting SMTs and show why they have been found attractive and plausible. I present the two challenges facing the application of SMTs to AVHs. I then introduce, motivate and clarify the PPF. I then present evidence suggesting that predictive processing might be disrupted in psychosis. Finally, I end by applying the PPF to voice-hearing, and show how it can, first, address the auditory phenomenology challenge, and second, nicely account for the three subtypes of AVH I present.
