Spatial working memory and problem solving in schizophrenia: the effect of symptom stabilization with atypical antipsychotic medication.

Reasoning and problem solving in the spatial domain are important aspects of executive function that are reliably impaired in schizophrenia, and the Groton Maze Learning Test(c) (GMLT) provides a valid measure of spatial working memory. In the current study, 34 patients with first-episode schizophrenia and 20 matched controls were assessed for baseline spatial working memory abilities using this hidden maze learning test. Approximately one month after baseline assessment, allowing for symptoms to stabilize in response to treatment with therapeutic doses of atypical antipsychotic medications for individuals with schizophrenia, all participants were again assessed with the GMLT. Prior to pharmacologic intervention, patients with schizophrenia showed significant impairments in performance of all aspects of the GMLT, including measures of learning efficiency and error monitoring. One month of treatment was associated with a reliable improvement in these domains, although impairments in accuracy and error monitoring on this spatial working memory test persisted despite symptomatic improvement. These results indicate that impairments in spatial working memory are present at the earliest stages of the illness, and that such deficits in performance remain present, albeit ameliorated, after treatment with atypical antipsychotic medication.