Depressive states are the predominant feature of bipolar disorder and a major source of morbidity and mortality in bipolar patients. Symptoms include affective changes, neurovegetative signs, and significant cognitive deficits. Several lines of evidence suggest that these symptoms are linked to neuropathological abnormalities in structures making up the anterior limbic network (ALN), a brain network hypothesized to modulate emotional homeostasis. These abnormalities are manifest by functional neuroimaging changes in regions involved in emotional control including the ventrolateral prefrontal cortex (VLPFC), anterior cingulate (ACC) and amygdala, as well as by neurochemical changes consistent with elevated neuronal metabolism. Magnetic resonance spectroscopy (MRS) studies of depressed bipolar patients suggest that abnormalities in energy metabolism may underlie increased prefrontal and medial temporal activity observed with functional magnetic resonance imaging (fMRI) and positron emission tomography (PET). Patients with bipolar disorder demonstrate significant changes in prefrontal concentrations of N-acetyl aspartate (NAA), as well as changes in phosphomonoesters (PME) and the presence of lactate, that suggest altered energy metabolism. Evidence of increased glutamate in the prefrontal cortex suggests that this increased neuronal activity is related to increased intraneuronal excitatory signaling within the prefrontal cortex. Together, these data support a model in which bipolar depression is marked by a loss of emotional modulation linked with increased ACC and VLPFC activation, and inhibition of, or interference with other brain regions, including the dorsolateral prefrontal cortex, which mediates cognitive domains impacted by bipolar disorder. These prefrontal changes in neuronal activity appear to be responsive to pharmacologic intervention. With these consideration in mind, the goals of this study are: 1) To use 1H-MRS to identify neurometabolic abnormalities in early-episode depressed bipolar patients, and to determine how abnormalities change in response to specific treatments; and 2) To identify corresponding changes in fMRI brain activation, in order to provide neurofunctional correlates to neurochemical markers of treatment response. To accomplish these aims, we will acquire integrated neurometabolic (MRS) and neurofunctional (fMRI) measurements in earlyepisode depressed bipolar patients and a matched cohort of healthy subjects in order to refine neurophysiological models of bipolar disorder (Center goal 1); to identify MRS and fMRI markers of treatment response in bipolar depression to two mechanistically different medications (Center goal 2); and to identify potential predictors of treatment response for future studies (Center goal 3).