Dementia is, of course, the label for a clinical syndrome with many different causes. Alzheimer's disease (AD) is the most common cause of dementia, followed by vascular dementia, dementia with Lewy bodies (DLB) and (especially at younger ages) frontotemporal dementia. Other possible causes of the dementia syndrome with relevance to a debate on the use of neuroimaging include space-occupying lesions, normal pressure hydrocephalus and other rarer neurological diseases which can be associated with variable cognitive impairments, such as progressive supranuclear palsy, multi-system atrophy and corticobasal degeneration. There is a great and, with the aging of the population, ever growing demand for the investigation of patients with suspected dementia, as exemplified by the rapid expansion of services such as memory clinics. It is essential that everyone presenting with signs suggestive of cognitive failure is given the opportunity to undergo a full assessment to help determine the subtype of dementia, which will include appropriate investigations. However, it is equally essential that investigations are appropriately targeted, according to their evidence base, and only performed when they have the potential to have a significant bearing on diagnosis or management.