Privileged Access and the Agent in the Thought-Insertion

In his paper, Young has eloquently put forward a novel account of how and why the phenomenon of thought-insertion (TI) seen in patients with schizophrenia does not contradict the immunity principle (i.e., first-person access to one's own thoughts and psychological states is 'immunity to error through misidentification'). He argues that, in TI, the problem lies not in misidentification but in mispredication: the individual with TI does not ascribe the right predicate to the wrong subject, but has misdetected the predicate in the first place. The author points out that an inconsistently formulated immunity principle could risk confusing the two types of errors. The author defines the immunity principle as:

Any agent who thinks a thought* (the asterisk indicates thoughts obtained by privileged access) of the form 'I am F' cannot be right on the basis of the perspective grounding this thought that something is F but wrong about whether 'I amF.'

(p. 153)

Given this definition, it would indeed be the case that, if one claims they have thoughts that are not theirs, it is the predicate and not the initiator of the thoughts that is susceptible to error. Note that the author stresses 'any formulation of the immunity principle be limited to present-tense self-referential judgments [my emphasis]' (p. 151) ascribing a predicate to the essential indexical 'I', and any over-simplification would only create the illusion that TI breaches the immunity principle.

However, perhaps another possibility for this confusion is the inconsistency in defining TI itself. The author of the current paper makes a specific point that, in TI, the patients do not only claim the thoughts are not generated by themselves, but actually belong to another agent. This, of course, would seem implausible (whether considered from the immunity principle or not) if not utterly false even to a lay person; and it is this apparent falseness that defines TI, at least in this instance, as a delusion. Indeed, throughout his paper Young has consistently referred to TI as 'delusions of thought-insertion' and implies that the delusion is primary in nature ('patients seem to require no external evidence, investigation or complex inference in order to arrive at them [the delusions],' p. 156). This conforms with the psychiatric definition of TI as a 'false belief that the subject receives inserted, alien thoughts' (Mullins & Spence, 2003).

Arguments about whether all delusions are beliefs aside, this definition could be divided into two components: the false belief and the inserted alien thoughts. Just by this medical definition alone, it is clear that the belief part has been dissociated from the actual inserted thoughts. So, if TI is indeed a delusion, to which of these two components does the delusion apply? It would seem obvious that the false belief about the inserted thoughts is the delusion. Some may argue that how could thoughts be inserted from the external (and also belong to someone else) in the first place? 'Privileged access' allows one to have direct, taken-for-granted knowledge (a kind of knowledge that does not require external evidence) of one's own thoughts, which are obtained through introspection. Yet, thoughts derived from both introspection and the mere act of thinking are separated from the individual's subjectivity in TI. This separation (some call it the 'separability thesis'; see Gibbs, 2000), I argue, is the experience of having seemingly external thoughts. The individual who has such an experience does not deny the fact that they are the one who has had external thoughts inserted into their mind; the individual simply denies the thoughts themselves are their own. Indeed, this would fit nicely with the author's account about mispredication. However, does this always constitute a delusion?

I have recently proposed (Humpston & Broome, 2015) that the experience of having external thoughts—whose externality is only accessible through the first-person perspective (which is again a paradox)—alone is insufficient for a delusional elaboration to form, and it is the act of ascribing another agent to these thoughts that drives the individual to a delusional endpoint of the phenomenon. I do not...