To the Editor: The article by David Sultzer, M.D., et al. (1) provides strong additional support, by way of correlation analyses of the observer-rated severity of delusions, for the contribution of right frontal brain dysfunction to the appearance of abnormal beliefs in Alzheimer’s disease. This form of analysis has the merit of accounting for the contribution of other variables, such as age, age at onset, and severity of dementia, as well as the behavioral factor of agitation, to variations in regional brain metabolism. There are, however, some comments to be made about the interpretation of the results and, perhaps more important, about the method of study adopted by the authors.

The findings were seen as evidence for a linear relationship between delusional "severity" and the degree of impairment of metabolism in areas of the right frontal cortex. There are challenges to this interpretation. It is equally possible that the content and personal significance of the delusions described (about half of those outlined could reasonably be considered elements of a misidentification syndrome) might have had some variable influence on the behavioral assessment of delusion severity on the Neurobehavioral Rating Scale. In other words, an association of the nature, as much as neuropsychiatric severity, of abnormal beliefs with quantitative variation in regional brain metabolism has not been fully examined. Equally, there is evidence from case studies that delusions that have a substantial impact on behavior (and would have been highly rated on the Neurobehavioral Rating Scale) may appear at the minimal stage of Alzheimer’s disease in association with subtle and confined cortical dysfunction and that they impair a specific set of cognitive abilities (2, 3).

The results of the study extend previous evidence from cross-sectional studies of similar populations. Reliance on a dimensional approach in a group showing diverse delusional phenomena, however, may continue to divert attention from methods more likely to foster an analytic understanding of delusional states. These methods will rely on the study of multiple single cases, as has been so fruitful in the analysis of Capgras syndrome (4), and will likely combine detailed clinical phenomenology, functional imaging, and cognitive neuropsychology (5). The discrimination of delusions with a factual content satisfying traditional clinical criteria from affectively laden persecutory beliefs may well be of heuristic value but will not sufficiently inform etiological studies in both organic and functional delusional disorders. Firmly held factual delusional beliefs can arise from specific memory failures and be affectively laden when the disorders of memory or other aspects of cognition involve issues of autobiographical knowledge and personal identity.