Verbal declarative memory is one of the most reliably impaired cognitive functions in schizophrenia. Important issues are whether the problem is reversible, and which brain regions underlie improvement. We showed previously that glucose administration improved declarative memory in patients with schizophrenia, and sought in this pilot study to identify whether glucose affects the location or degree of activation of brain regions involved in a verbal encoding task. Seven clinically stable and medicated patients with schizophrenia or schizoaffective disorder, who showed deficits on a clinical test of memory, participated in the study. Subjects served as their own controls in a double-blind, crossover protocol that consisted of two sessions about a week apart. In each session, subjects ingested a beverage flavored with lemonade that contained 50 g of glucose on one occasion, and saccharin on the other. Blood glucose was measured before and 15, 50, and 75 min after ingestion. After ingesting the beverage, they performed a verbal encoding task while undergoing brain functional magnetic resonance imaging. The results showed significantly greater activation of the left parahippocampus during novel sentence encoding in the glucose condition, compared to the saccharin condition, despite no change in memory performance. A trend towards greater activation of the left dorsolateral prefrontal cortex (p<.07) was also evident in the glucose condition. These pilot findings emphasize the sensitivity of both the medial temporal and prefrontal regions to effects of glucose administration during encoding, and are consistent with the hypothesis that these regions also participate in declarative memory improvements following glucose administration.