It’s not that these studies don’t show any predictive value of brain scans, but so far, none have proved reliable enough to be even slightly useful in clinical practice.
It’s also likely that brain scans will never predict schizophrenia 100%, because the evidence suggests that schizophrenia isn’t a single cut-and-dry condition, like the flu
or arthritis – it’s just a vaguely described set of experiences that tend to occur together and often cause the person distress or impairment.
In fact, two people could have schizophrenia and not share any symptoms. In other words, there’s little that’s definitive and definite to predict.
Now, if only brain scans could predict when journalists will write a ‘brain scans predict’ article, then we might be able to intervene with the appropriate help.
UPDATE: There’s some good comments on this post, and CopperKettle has provided a link to the full paper.
Link to BBC News story “Scans ‘can predict schizophrenia'”.