Read on Twitter
Terrific paper. I am largely in agreement with the arguments made. A few thoughts: 1. If cognitive testing deficits are related to epiphenomena, they are still important from a clinical perspective even if they are not 'directly' related to the illness https://twitter.com/Dr_Rebecca_D/status/1304420741237137408
2. We could examine this with better efforts to measure 'everyday cognition'. Again, high correlations between cognition and how well people with schizophrenia are functioning suggests that the tests we use have good predictive validity.
3. I think the argument does overlook to some extent the degree of impairment before illness onset, but these studies are very difficult to do and there are likely many biases with trying to retroactively assess cognition and sampling issues with prospective studies
4. There is evidence that when cognitive performance is in the average range compared to the general population, there is still a 'deficit' if compared to the individual's expected performance (relative to premorbid or MZ twin performance).
5. Especially in the early stages of psychosis, where there is so much diagnostic uncertainty, we need to be careful because many will not eventually meet criteria for schizophrenia and likely including them in group will inflate the mean performance.
6. I love the idea of trying to disambiguate 'true' cognitive impairment from the degree to which performance is related to these additional factors the authors mention. They have outlined important considerations and many are testable hypotheses.
