STANDARDIZED VARIMAX DESCRIPTORS OF EVENT-RELATED POTENTIALS - EVALUATION OF PSYCHIATRIC-PATIENTS

A large normative data base of visual and auditory event related poten tials (ERPs) was collected. Factor analysis (PCVA) was used to extract factor wave shapes that accurately reconstructed these normal ERPs wi th appropriate ''factor scores.'' The mean value and standard deviatio n (SD) of the normative factor score distribution were computed separa tely for each stimulus, factor, and electrode. This enabled reconstruc tion of any individual ERP as a combination of these standardized Vari max descriptors, with z-transformation of the required factor scores g iving objective statistical assessment of ERP wave shape. Statistical probability factor z-score topographic maps were constructed, color co ded in SDs from the normative means. The incidence of significant indi vidual deviations from these normative mean values was at or near chan ce levels in test groups of normal subjects. For many of these new ERP descriptors, significant deviations from the norms were found for sin gle features in from 20% to as much as 63% of the patients in particul ar diagnostic categories. Factor z-scores were used to construct multi variate discriminant functions that accurately and replicably separate d (1) normal from schizophrenic from demented subjects and (2) schizop hrenic from bipolar depressed subjects.