Most, but not all, persons diagnosed with schizophrenia are able to learn when given systematic, repetitive exposure to a perceptual or motor task. In spite of their normal skill acquisition, persons with thought disorder or reality distortion might not be able to learn by using normal brain physiology. The behavioral and physiological impact of particular symptom profiles on perceptual learning has not been assessed in an experimental design explicitly created to generate similar performance across subjects of differing abilities. This proposal will use event related functional magnetic resonance imaging (fMRI) with normal healthy volunteers and persons with schizophrenia, to determine whether thought disorder (disorganization), or reality distortion (hallucinations plus delusions) specifically influence the physiological response to learning a visual / spatial recognition task. We expect thought disorder to have a significantly greater effect on a schizophrenic's neural activity patterns associated with learning, than reality distortion. Thought disorder is hypothesized to diminish hippocampal and frontal cortex response to training, especially during the encode phase of the trial. We will determine the dynamic roles of particular cortical systems in normal subjects before and after visual recognition training. We will then assess the relative impact of these two symptom profiles, in persons with schizophrenia learning a visual Delayed Match to Sample Task (DMST), on the changes found in these regions. We will obtain event related fMRI studies on 30 clinically stable, medicated persons with schizophrenia (15 with thought disorder and 15 with reality distortion) and 25 age and sex matched comparison subjects (normal volunteers) before (first fMRI) and after (second fMRI) visual skill training using a visual DMST. Clinical characteristics, thought disorder (TD) and reality distortion (RD), of the schizophrenic volunteers will be correlated with the physiological (especially encode components of the trial) and behavioral response to visual learning. A third fMRI study after 8 weeks of intensive visual training, will be done on all schizophrenic volunteers following the second study. SZ with predominantly RD may be able to use this extended training to further normalize their task-activated BOLD signal patterns but SZ with predominantly TD may not benefit. Our understanding of how symptom clusters contribute to abnormal physiological activity patterns associated with learning visual recognition.