Intact and impaired cognitive-control processes in schizophrenia

Deficits of cognitive-control in schizophrenia have been assumed to result from a single impairment that leads to widespread consequences. Contrary to this view, we hypothesized that different control processes operate at different stages of processing, and that only some of these processes may be impaired. We employed two selection tasks to test the hypothesis that patients with schizophrenia have deficits in selecting information in working memory (WM), but not in selecting perceptual information. In the “Ignore” task, which fosters perceptual selection, participants saw a cue to remember either red or blue words, followed by a memory-set (2 red, 2 blue), a brief delay, and then a probe. The “Suppress” task was similar, except the memory-set came before the instruction-cue, and hence selection had to occur in WM. We recorded reaction time and percentage errors for positive probes (“Valid”), and two kinds of negative probes, those that were supposed to have been dropped from WM (“Lures”) and those that had not appeared in the memory-set (“Controls”). Compared to healthy controls, patients were impaired in the Suppress but not the Ignore task. This dissociation implies that there are two different selection mechanisms.