Philosophy, Psychiatry, & Psychology

Philosophy, Psychiatry, & Psychology 9.1 (2002) 61-63

Arguing from Neuroscience in Psychiatry

James Phillips

PHILIP GERRANS "A One-stage Explanation of the Cotard Delusion" provides an elegant example of the application of neuroscientific findings to known clinical phenomena in psychiatry. Gerrans argues that, in the cases of the Cotard and Capgras delusions, a one-stage explanation is sufficient to account for the clinical phenomena. That is, the neuroanatomic defect and the consequent anomalous experience are sufficient to produce the delusion. According to Gerrans, we do not need to postulate that the delusion will emerge only if the anomalous experience is filtered through a faulty reasoning process (a two-stage explanation). Against the argument of Young and Leafhead (1996) that Cotard and Capgras delusions stem from the same neuroanatomic deficit and differentiate only in being processed differently—thus requiring a two-stage explanation—Gerrans argues that the respective neuroanatomic defects are in fact different, and that, therefore, the two delusional syndromes differ at the first stage and do not require the two-stage explanation suggested by Young and Leafhead.

I do not have an opinion as to who is right in this disagreement, and I will devote these remarks to explaining why I do not have an opinion. In developing this explanation I will touch on what I find problematic in many discussions of neuroscience, neurocognitive models, and clinical psychiatry; I will thus use this article to highlight those problems.

On the face of it, one would think that the issue raised by Gerrans should admit of a simple resolution. He argues that Capgras and Cotard delusions derive from different neural defects. Young and Leafhead, on the other hand, base their two-stage argument on the argument that the two delusions derive from the same neural defect. So which is it? Neuroscientists should be able to settle the issue and end the dispute. Gerrans remarks "that we need to look more closely at the CN [cognitive neuropsychiatric] explanation of the Capgras delusion." As he describes it, face recognition involves two components, a cognitive and an affective component, with separate neuroanatomic substrates. An individual successfully recognizes the face of a friend or acquaintance when she matches the cognitive recognition (cognitive component) with the affective tone of familiarity (second component). When there is a lesion in the neural substrate of the affective component, she may recognize the face but without the tone of familiarity. In her effort to understand her affective experience of the face as alien rather than as familiar, she may conclude that her friend has been replaced by a double. Thus we have the Capgras delusion.

Let us now unpack this explanation. According to Gerrans, researchers have hypothesized [End Page 61] two anatomatically distinct pathways involved in face recognition. He writes: "Initially . . . these separate neuroanatomic substrates were hypothesized to lie in separate dorsal and ventral processing pathways, but the most persuasive hypothesis is that of Breen et al. that face recognition is performed in the ventral visual processing pathway, whereas covert [affective] recognition effects result from the activation of pathways connecting the ventral pathway and the limbic system. It is damage to these latter pathways that gives rise to the Capgras delusion."

In other words, the explanation of Capgras delusions depends on a neurocognitive model, itself based on an unspecified base of neuroanatomic data. When Gerrans writes that "[t]he Capgras delusion arises in cases where, following a neurological accident, the pathways connecting the FRU [Face Recognition Unit] to the limbic system are damaged leaving ventral pathways and hence overt recognition unimpaired . . . ," he means that the Capgras delusion arises in that manner if the proposed model bears any relation to the reality of the living brain and if real Capgras patients have suffered the hypothesized lesion. At this point the reader is struggling with the following questions: (1) What are the hard neuroanatomic data underpinning the neurocognitive model? (2) What is the argument for developing this model (single Face Recognition Unit with independent pathways to cognitive and affective components of face recognition) out of the hard data? (3) What is the evidence...