Brain morphometry and functional imaging techniques in dementia: methods, findings and relevance in forensic...

The ability to predict what people perceive from patterns of brain activity raises futuristic questions. Amongst these are questions about the role of brain activity in predicting misdemeanours or preventing them. Two obvious cases in point are the tendency of some patients with fronto-temporal lobar degeneration to become aggressive and the difficulty deciding when Alzheimer patients need to stop driving for reasons of safety. These two situations will be used to structure a review of the literature in this general area.

Multivariate classification techniques (MCT) improve early accurate diagnosis of dementia. Given the known frequency of behavioural abnormalities, this information allows better prediction of the future frequency of such behaviour. In addition, MCT could prove suitable for providing reasonably accurate information of relevance to individuals about the combination of future symptoms. However, no study has applied MCT to the prediction of future behavioural problems or to assessments of road safety in dementia.

MCT could improve the prediction of offensive or risky behaviour in which standard neuropsychological testing is less than conclusive. Cognitive function in multiple domains, as required for driving, is likely to be best examined using well established neuropsychological tests and possibly driving simulators.