Neuropsychological research yields diverging results regarding Working Memory (WM) in Obsessive-Compulsive Disorder (OCD). In the present study an attempt was made to focus in the differences between OCD patients and healthy controls, as reflected by the P600 component of ERP signals, as well to search deeper into the P600 signals by extracting new features and, by employing powerful classification procedures, to develop a pattern recognition system for discriminating OCD patients from controls. Eighteen patients with OCD symptomatology and twenty age and sex matched normal controls were examined. All subjects were evaluated by a computerized version of the digit span subtest of the Wechsler Adult Intelligence Scale. EEG activity was recorded from 15 scalp electrodes (leads). From the P600 component of each signal nineteen waveform-features were calculated. The Probabilistic Neural Network (PNN) classifier was developed and it was fed with features from all leads. Highest single-lead precision (86.8%) was found at the Fp2 and C6 leads. When leads were grouped into anatomical regions, highest accuracies were achieved at the temporo-central (86.8%) region (C5,C6). These findings may be indicative that OCD patients present deficits related to WM mechanisms, corresponding to prefrontal, central, and temporocentral regions, as reflected by the P600 component.