The election seems to be mad, and instead of concentrating on that I’m reading journals and doing work. This is in JAMA Psychiatry pre publication stack. It is a fairly well done case control study. The associations they find are descriptive and the links causal.
But the note that the issues relate to memory and not verbal assessment do shift paradigms.

Fhe unexpectedly higher verbal score (reflecting WRAT-4 reading) that was retained in logistic and Cox proportional hazards regressions in concert with impaired declarative memory abilities was not a significant predictor in univariate comparisons. This pattern of high verbal premorbid ability and impaired memory, coupled with the unusual thought content or delusional ideas and suspiciousness or persecutory ideas items composite, appears to be a pernicious combination predicting conversion and needs replication. Also important, the BVMT-R (a visual-memory test) showed comparably large impairments as the 2 verbal memory tasks, highlighting that deficits of declarative memory abilities in CHR are not solely verbal and that declarative memory abilities impairments are key neurocognitive risk markers.

Neurocognitive tests used in concert with other clinical and psychobiological measures may enhance prediction of psychosis or functional outcome. For example, in analyses limited to 2 tests selected from a literature review14 before these neuropsychological analyses, the NAPLS 2 investigators found that the HVLT-R and the Brief Assessment of Cognition in Schizophrenia symbol coding tests added modest but significant independent predictive power above the clinical measures in a risk calculator algorithm for psychosis conversion,45 and this finding was replicated in an independent non-NAPLS 2 sample.

Why could they shift paradigms? One of the core symptoms of psychosis is formal thought disorder. This is apparantly a verbal deficit, but as the video notes, it relates to what Karl Jaspers calls un understandability. IT is not an aphasia: the word salad for that is different.

Our approach to diagnosis is to look for signs in an interview: often these come when people are not talking about their difficulties but general life. We listen as much to how things are said as what is said. We don’t usually do neuropsychological batteries. These are time consuming and expensive. The correlations are not (yet) sufficient to allow for causality or significant prediction to intervene with the knowledge that this will stop the progression from difficulty to madness.

However, if the current theories that madness is a disorder of neural networks is correct, this would fit. Another area to watch carefully.