Currently, we use quantitative analysis of spontaneous (EEG) and event-related potentials (ERPs), to develop 1) brain images of working memory (WM) and 2) biological classification of psychiatric patients. 1) We measure WM via the presentation of information items, e.g., faces, letters, and numbers, in a Rpriming setS (S1) followed by a Rmatching setS (S2) of items that must be compared with the previous sample. Items in sets S1 and S2 are flashed on a screen at a specific rate and Rprobe stimuliS consisting of blank flashes are flashed at the same rate during intervals before and after S1 and S2. We use very narrow band spectral analysis of the EEG and principal component analysis of ERPs to examine voltage fields or RlandscapesS on the head during successive intervals of these delayed matches from sample tasks. Landscapes to probe stimuli after S1 differ from those before, reflecting the engagement of an extensive neuronal population in mediating WM of S1. Analysis of the Rdifference landscapeS shows that it is a persisting image of the voltage fields elicited by both S1 and S2, varying as a function of the psychiatric patients have been subjected to cluster analysis based on EEG and ERP measurements evaluated relative to a normative data base. Preliminary results indicate that patients with different neurometric profiles respond to different pharmacotherapeutic treatments. Large-scale international collaborative studies are being organized to collect a patient cohort of sufficient size for prospective confirmation of using predrug neurometric evaluations to predict selective treatment outcomes.