A post by Kengo Miyazono.

Normally we have no difficulty in distinguishing what we believe from what we imagine. We seem to have a reliable metacognitive capacity that enables us to distinguish our beliefs from our imaginings. I can easily judge that “the university library is closed today” is something I believe and that “I am the best football player in the world” is something I imagine. But how exactly do I do this? How exactly do I distinguish beliefs from imaginings?

Several psychologists (e.g., Wegner 2017, Wilson 2004) and empirically-oriented philosophers (e.g., Carruthers 2009, 2011) argue that self-knowledge is achieved in the process of self-interpretation; i.e., the process of observing and interpreting own behavior. For example, I come to know that I believe that the university library is closed today by observing and interpreting my own behavior. This process is the same in kind as the process in which I come to know that somebody else, John, believes that the university library is closed today. I come to know it by observing and interpreting his behavior. Let us call this view “Self-Interpretation Theory”. (Strictly speaking, Self-Interpretation Theory comes in different forms, but here I focus on the most simple version of it.)

Self-Interpretation Theory answers our initial question as follows; we distinguish beliefs from imaginings by observing their behavioral consequences. The belief that P and the imagining that P have different behavioral consequences, and this behavioral difference enables us to tell whether the mental state at issue is a belief or an imagining. For example, I judge that I believe that the university library is closed today by observing that I behave as if I believe that the university library is closed today  (e.g., I do not visit the university library today). I judge that I imagine that I am the best football player in the world by observing that I behave as if I imagine that I am the best football player in the world (e.g., I pretend that I am the best football player in the world).

In this post I raise an objection to Self-Interpretation Theory; this theory seems to have difficulty in explaining the metacognitive beliefs of delusional patients. 

On one hand, delusional patients often attribute their delusions to themselves as their beliefs. This suggests that delusional patients regard their delusions to be beliefs. In other words, a patient with the delusion that P believes that he believes that P. A patient with the Capgras delusion that his wife has been replaced by an imposter, for example, believes that he believes that his wife has been replaced by an imposter. I call this fact about delusions “Doxastic Recognition”.

Note that Doxastic Recognition does not imply that delusions are in fact beliefs. It only says that at least delusional patients themselves regard their delusions as beliefs. Currie and colleagues (Currie 2000; Currie & Jureidini 2001; Currie & Ravenscroft 2002), who claim that delusions are not beliefs but imaginings, admit that delusional patients (mistakenly) regard their delusions as beliefs.

On the other hand, delusions often fail to have belief-like impact on behavior. The Capgras patient, for example, might fail to behave as if he believes that his wife has been replaced by an impostor (e.g., visit a police station to ask for help). Stone and Young write: “[A]lthough in some cases of Capgras delusion patients act in ways that seem appropriate to their beliefs, in many other cases one finds a curious asynchrony between the firmly stated delusional belief and actions one might reasonably expect to have followed from it” (Stone & Young 1997, 334). I call this fact about delusions “Lack of Action Guidance”.

It is not very difficult to see that Self-Interpretation Theory faces a problem in accounting for the metacognitive beliefs of delusional patients. Given the lack of relevant behavior (Lack of Action Guidance), the theory cannot explain why delusional patients regard their delusions as beliefs (Doxastic Recognition). The theory seems to predict that delusional patients, who do not exhibit relevant behavior, do not regard their delusions as beliefs. It seems to predict, for instance, that the Capgras patient, who does not behave as if he believes that his wife has been replaced by an impostor, does not regard his delusion as a belief.   

Currie and colleagues suggest that functional roles (including the functional roles concerning behavioral outputs) of delusions are similar to those of typical imaginings (possibly, more similar to them than to the functional roles of typical beliefs). If this is true, then Self-Interpretation Theory faces an even deeper problem; the theory would predict that the Capgras patient, whose behavioral outputs are similar to the behavioral consequences of imaginings, regards his delusion as an imagining. Egan (2009), in contrast, argues that functional roles of delusions are intermediate between those of typical imaginings and those of typical beliefs. In that case Self-Interpretation Theory might predict that the Capgras patient, whose behavioral outputs are intermediate, regards his delusion as an intermediate state (“bimagination” in Egan’s term). Either way, Self-Interpretation Theory fails to explain why the Capgras patient regards his delusion as a belief rather than an imagining or an intermediate state.

This problem seems to show that Self-Interpretation Theory, at least in its simplistic form, is untenable. What should we do, then?

An option is to revise Self-Interpretation Theory. For example, one might weaken the theory by abandoning the idea that self-interpretation is the only process of self-knowledge and recognizing another process; i.e., a direct and introspective access to own mental states. According to Bem’s Self Perception Theory, which is an example of this option, we rely on the process of self-interpretation only when the direct and introspective access is not very helpful; “to the extent that internal cues are weak, ambiguous, or uninterpretable, the individual is functionally in the same position as an outside observer, an observer who must necessarily rely upon those same external cues to infer the individual's inner states” (Bem 1972, 2). This is certainly a possible view, but we still need to know the nature of the “direct and introspective access” through which delusional patients judge their delusions as beliefs. 

Another option is to keep the idea that self-interpretation is the only route to self-knowledge but adopt a broader conception of “self-interpretation”. According to Carruthers’ (2011) Interpretive Sensory-Access Theory, which is a good example of this option, the materials for self-interpretation include not only the external behavioral cues but also inner sensory cues such as inner speech and visual imagery.

According to Interpretive Sensory-Access Theory, the Capgras patient identifies his delusion as a belief rather than an imagining (or something else) by observing and interpreting inner sensory cues. However, inner sensory cues might not be informative for the purpose of distinguishing beliefs from imaginings. For instance, can we distinguish, by observing inner speech, the belief that my wife has been replaced by an impostor from the imagining that my wife has been replaced by an impostor? Probably, both in believing and imagining, I find in myself the same inner speech sentence “my wife has been replaced by an impostor”. But then this inner speech sentence tells us nothing about whether “my wife has been replaced by an impostor” is something I believe or I imagine.

Can visual imagery be more useful than inner speech? Can we distinguish, by observing visual imagery, the belief that my wife has been replaced by an impostor from the imagining that my wife has been replaced by an impostor? How? Is the imagery more vivid in the case of believing than in the case of imagining? (Hume has a similar proposal in his A Treatise of Human Nature, but it has been heavily criticized by commentators; e.g., Bennett [1971], Stroud [1977].) 

The last option is to explore an alternative account, which does not appeal to the idea of self-interpretation. An example is what I call “Phenomenal Access Theory” according to which we distinguish beliefs from imaginings by the phenomenal difference between them. The belief that P and the imagining that P have different phenomenal characters, and this phenomenal difference enables us to tell whether the mental state at issue is a belief or an imagining (e.g., Goldman 1993). According to this proposal, delusional patients regard their delusions as beliefs because the delusions have doxastic phenomenal character.

We need more discussions to see which is the best option. My claim here is that the correct theory of self-knowledge, whatever it turns out to be, needs to be able to explain the peculiar facts about delusional patients and their metacognitive beliefs.

References

Bem, D. (1972) Self-perception theory. in Advances in Experimental Social Psychology, vol. 6, L. Berkowitz (ed.). Academic Press, 2-62.

Bennett, J. (1971). Locke, Berkeley, Hume: Central Themes. Oxford University Press.

Currie, G. (2000). Imagination, delusion and hallucinations. Mind & language, 15(1), 168-183.

Currie, G., & Jureidini, J. (2001). Delusion, rationality, empathy: Commentary on Martin Davies et al. Philosophy, Psychiatry, & Psychology, 8(2), 159-162.

Currie, G., & Ravenscroft, I. (2002). Recreative Minds: Imagination in Philosophy and Psychology. Oxford University Press.

Carruthers, P. (2009). How we know our own minds: The relationship between mindreading and metacognition. Behavioral and Brain Sciences, 32(2), 121-138.

Carruthers, P. (2011). The Opacity of Mind: An Integrative Theory of Self-Knowledge. Oxford University Press.

Egan, A. (2009). Imagination, delusion, and self-deception. in Delusions and Self-deception; Motivational and Affective Influences on Belief Formation, T. Bayne & J. Fernández (eds.). Psychology Press, 263-280.

Goldman, A. I. (1993). The psychology of folk psychology. Behavioral and Brain Sciences, 16(1), 15-28.

Stone, T., & Young, A. W. (1997). Delusions and brain injury: The philosophy and psychology of belief. Mind & Language, 12(3‐4), 327-364.

Stroud, B. (1977). Hume. Routledge.

Wegner, D. M. (2017). The Illusion of Conscious Will. New Edition, MIT Press.

Wilson, T. D. (2004). Strangers to Ourselves. Harvard University Press.