- Initially, Paul Grant and I noted that the apparent decrements on neurocognitive tests (attention, recall, flexibility, and social cognition) did not correspond to the symptomatology of schizophrenia. Since it had already been established that poor performance on the tests did not correlate with positive symptoms, then we had to explain the fact that the test performance did correlate with negative symptoms.
- It occurred to us that defeatist and asocial attitudes might mediate this correlation. We confirmed that defeatist attitudes did mediate between the two. Six other investigators confirmed this finding.
- In a longitudinal study, we found that dysfunctional attitudes and asocial attitudes had also mediated but had separate pathways to the negative symptoms.
- Robyn conducted a review of the literature and showed that some confounds influenced the relationship between the neurocognitive test scores and negative symptoms (and other variables).
- Also, we noted from the literature that the items of interest, namely, attention, etc. became the object of focus because these were found to show defective functioning in brain-damaged individuals. We propose that brain-damaged individuals were an inappropriate comparison group.
- Looking over the literature on test taking, it seemed that the individuals did poorly on whatever test they took. Also, poor performance on these tests could be improved through specific strategies (see Michenbaum; Proverbs Test paper from Harvard).
- A generalized finding would be that these individuals are task aversive. This is demonstrated in the paper by Moritz, which found that negative test-taking attitudes preceding the administration of the test, and then following the test showed that these attitudes (which were a conglomeration of the so-called confounds) were an influence on test performance.
- Conclusion: the tests are not measuring the appropriate variables; the search for the neurocognitive problem in schizophrenia should look at other factors that are more central to the disturbance, than that tested in the neurocognitive tests. Ideally, tests should be developed for identifying these impairments and, they should correlate with positive, as well as negative symptoms. Also, the tests should include a focus on the individual’s positive assets, goals, and motivations, which may play a greater role in recovery than to the deficits. Also, in so far as possible, a new test should be decontextualized, perhaps using something like the dynamic testing method.