A failure of sleep-dependent procedural learning in chronic, medicated schizophrenia

Background Schizophrenia patients have difficulty mastering even rote procedural tasks in rehabilitation settings. Although most studies demonstrate intact procedural learning in schizophrenia, recent findings demonstrate that a critical component of procedural learning is dependent on sleep. This study tested the hypothesis that patients with schizophrenia have a deficit in sleep-dependent procedural learning. Methods Using a simple, well-characterized test of motor skill learning, the finger tapping motor sequence task (MST), 26 patients with chronic, medicated schizophrenia and 14 demographically matched healthy control subjects were tested on two occasions, 24 hours apart. The main outcome measures were learning of the MST on day 1 (practice-dependent learning) and overnight, sleep-dependent improvement in performance. Results Although schizophrenia patients and control subjects did not differ in practice-dependent learning, patients failed to show overnight improvement (4% deterioration) and differed significantly from control subjects who showed a significant 11% improvement. Conclusions We present here the first demonstration of a failure of sleep-dependent consolidation of procedural learning in chronic, medicated schizophrenia. This deficit occurred in the context of normal practice-dependent learning within a training session. This behavioral dissociation is consistent with evidence that practice- and sleep-dependent motor learning reflect independent processes and suggests that they are differentially affected in schizophrenia.