However, an alternative explanation may also arise. In a PET study (3), McGuire et al. found decreased activation in the left rostral supplementary motor area in hallucinating schizophrenic patients when they were requested to imagine words spoken in another person’s voice. This task involved both the generation and monitoring of internal verbal activity. McGuire et al. concluded that the decreased activation in the supplementary motor area served as a neural basis of deficient self-monitoring in these patients. Therefore, it is possible that in the word list recall task of Dr. Crespo-Facorro et al. (1), the underactivation of the supplementary motor area was related to impairment of the self-monitoring of inner speech. However, on the basis of published data, this question remains unresolved. It would be interesting to know whether their patient group included hallucinating schizophrenic subjects and, if so, how their brain activation patterns differed from those of nonhallucinating patients. Moreover, it could be that during retrieval, the patients imagined the words being said in the experimenter’s voice (the list was read to the subjects immediately before scanning) (1). These questions are important since, to our knowledge, independent research groups have not yet replicated the original findings of McGuire et al. (3).