Positron emission tomography with 18F-2-deoxyglucose (FDG) will be used to explore regional brain metabolic rate differences first in normal subjects during rapid eye movement (REM) sleep, non-REM (NREM) sleep, and waking. Thirty-six subjects, three groups of twelve each, will have PET scans after receiving FDG during polygraphically monitored sleep or a waking eye movement task. EEG and EOG monitored to classify sleep stage and to rigorously exclude FDG uptake periods not meeting criteria for duration and percent time in a stage. PET images would be converted to metabolic rate and mean rates within cortex and stereotactically located areas measured. Hypotheses of regional brain activation in REM sleep, similarity of REM and wakefullness, hemispheric asymmetry in REM sleep, and correspondence of primary sensory cortex activation and dream content will be tested by repeated measures ANOVA. PET images will be reconstructed into lateral cortical surface topographic maps for surface analyses. Trend surface and regional principal components strategies would be developed. High resolution PET scans, waking normal control data for other tasks and prescreening for specific sleep architecture are special features of this proposal. We plan on a second phase beginning with a renewal for year three of this project to study patients with affective disorders.