On abductive inference and delusional belief: Why there is still a role for patient experience within explanations of Capgras delusion

Abstract

This paper aims to examine critically the explanatory model of delusional belief presented in Coltheart, Menzies, and Sutton's (2010) paper, “Abductive Inference and Delusional Belief”. The authors acknowledge that certain aspects of the model are speculative. In return, I speculate over the likelihood that the model's emphasis on subpersonal processing adequately and coherently explains the symptoms (as best we know them) of patients with delusional misidentification (specifically, the Capgras delusion) and nondeluded equivalent patient groups. In addition, I offer an account of the Capgras delusion that is compatible with many of the tenets of Coltheart et al.'s model, but which preserves an important explanatory role for patient experience absent, and erroneously so, I contend, from the aforementioned model. The more integrated explanation I am proposing here also provides a number of pertinent empirical questions and testable hypotheses that could inform future models of delusional belief.

Keywords:

• Delusional misidentification
• Estrangement
• Misplaced being
• Unbidden thoughts
• Reduced SCR

Notes

¹In more recent times, the Capgras delusion has typically been considered to involve a deficit in the face recognition system, and examples have tended to refer to cases where a patient does not recognise the significant other (family member/friend, etc.) when face-to-face, an event that co-occurs with reduced SCR. There are exceptions to this, however. Lewis, Sherwood, Moselhy, and Ellis (2001), for example, report a case of reduced SCR to famous voices in a normal-sighted patient who nevertheless demonstrated normal SCR for faces. Reid, Young, and Hellawell (1993) likewise report a case of “Capgras for voices” in a blind patient.

² It is important to note that Coltheart et al. do accept that there are some forms of delusional disorder (e.g., delusions of reference and persecution) that may have an antecedent abnormality in the form of “a conscious experience” (2010, p. 265) rather than exclusively abnormal data processed subpersonally, as is claimed to be the case with the Capgras delusion.

³I thank the anonymous reviewer for drawing my attention to these comparisons.

⁴As noted earlier, the term “subpersonal” refers to that which is unconscious. However, to clarify further, in the context of this paper, a subpersonal process can be either neurological or cognitive. What makes the process subpersonal is simply that it occurs below the level of consciousness.

⁵In support of this discrepancy, Brighetti et al. (2007) found that Capgras patient YY demonstrated less eye fixation when viewing familiar faces (compared to matched unfamiliar faces), concluding that “identity recognition of familiar faces, associated with a lack of SCR, results in gaze avoidance of the eye region (p. 196).

⁶The emergence of the move towards unconscious processing can be seen in an earlier paper—Coltheart (2005).

⁷The word “tartling”, we are told by Cleary and Specker, has its roots in the Scottish language.

⁸Cleary and Specker (2007) likewise presented a series of faces to participants and asked them to judge which had been presented previously using two types of item-based recognition. They found that participants recognised previously studied items either by directly recalling that the face had been previously studied (recognition through direct recall), or “on the basis of a sense of familiarity with the test item” (p. 1610)—the stronger the sense of familiarity, the higher the rating given. Previously studied items tended to be rated as more familiar than unstudied items, even when neither could be directly recalled as having been studied or not.

⁹It could be argued that the prosopagnosic patient is making nonrandom choices similar to those made by blindsight patient, who can detect stimuli in the blind region of their visual field without any conscious experience of it being like anything to do so. By way of a response, I would argue that, in the case of blindsight, the patient believes, initially at least, that the choices are random—hence their surprise at their success. However, there is no indication that the prosopagnosic patient was simply making random choices. The task was not to randomly select one of the two faces (for which the patient then “randomly” selected the previously studies face above the chance level); rather, the patient was asked to select the preferred face. For the task to be meaningful as a preference task, I contend, the choice had not to appear random to the patient, but instead to be based on a sense of preference. The meaningfulness of the task (to the patient) is what distinguishes this task from the blindsight example. There is nothing in the original research to indicate the prosopagnosic patient believed he was choosing randomly, although I accept that it is a possibility. However, in the absence of any evidence of randomness, and in light of the assumed need for the patient to engage in a meaningful (nonrandom) way with the task, it is reasonable, I feel, to consider that there was a genuine attempt on the part of the patient to select based on a sense of preference—something that would, for the patient at least, make it seem like it was not just random selection, and that he was doing as instructed.

¹⁰Denburg, Jones, and Tranel (2009), for example, measured an increase in SCR in a patient suffering from simultanagnosia. The increase corresponded to the presentation of negatively charged stimuli (e.g., an image of a burned and bloodied body), despite the patient being unable to consciously identify what the image was. In this context, the large amplitude SCR was interpreted as the covert recognition of the (negative) affective valance of the image.

¹¹Ideally, it would have been useful to ask the patient how he was making his choice. However, in the absence of such evidence, I nevertheless argue that, in contrast to blindsight patients who (the literature informs us) consider the task of identifying a given stimulus to be initially pointless—because they can't “see” it, and therefore feel that all they can do is guess—the prosopagnosic patient engaged in a seemingly more meaningful task. To explain: in contrast, to the blindsight patient's task of identifying an object or the direction of its movement, having to state a preference does not require one to guess—in fact, the notion of having to guess seems inappropriate. If asked: “Which face do you prefer?” would the patient think “I'll take a guess and say I prefer this one”? Even if the patient had no strong preference for either face, the task of choosing is not based on a guess. It maybe that the choice is arbitrary but the fact that the original study did not report that the patient felt that he had no preference either way and was just making arbitrary choices is suggestive of the fact that the task was more meaningful to the patient than that carried out by the blindsight patient. It is suggestive, I contend, of the choice being more legitimately based on a sense of preference, however mild it may have been in some if not all cases.