It's not clear how useful this is as a diagnostic tool when the differences they found are well within the range of normal variation. If you have a patient who has a fluid intelligence score 0.5 standard deviations below average, is that a person of formerly average intelligence who's in the early stages of cognitive decline, or a person who was always a bit below average? Or maybe a person who started out well above average and has undergone severe decline?

I think you'd have to test patients multiple times over a period of years to get good diagnostic information out of this.

Also, does anyone know what the scale of the fluid intelligence score is? Like what's the mean and standard deviation for control subjects? Eyeballing Figure 1, it looks like the standard deviation might be two points?