The “for debate” series in International Psychogeriatrics offers an opportunity for different points of view to be put forward regarding a particular topic or controversial area of practice in our field. The contributions here are related to the use of neuroimaging in the investigation of those with suspected dementia. Advocates of neuroimaging suggest that it is a vital component in the workup of all those with suspected cognitive impairment, and this view has largely been reflected in guidelines such as those of the American Academy of Neurology (Knopman et al., 2001), the Royal College of Psychiatrists, the National Institute for Health and Clinical Excellence (NICE) in the U.K. (www.nice.ac.uk), and other eminent international groups including the European Federation of Neurology (Waldemar et al., 2000; Waldemar et al., in press). However, critics of this approach argue that neuroimaging represents an intensive resource which could be better used elsewhere, that the “added value” of improving diagnostic accuracy over and above that possible without neuroimaging remains to be proven (and, even more so, the cost effectiveness of such), and that use of neuroimaging should be selective and directed only to where it is likely to make a substantial impact on patient management.