Objective: It has been hypothesized that an unstable arousal system exists in patients with obsessivecompulsive disorder (OCD) and P300 is commonly used to investigate for such information-processing abnormalities. However, studies of P300 in OCD have had inconclusive results, regarding both amplitudes and latencies, along with poor correlation with clinical variables. This study aimed to further investigate the suggested misallocation of cognitive resources and attentional disorder in OCD. Methodology: Thirty patients of OCD, taken from the outpatient department of Central Institute of Psychiatry, and 27 age-, sex-, and education-matched normal volunteers were examined in this study. The psychopathology in the patients was evaluated using the Y-BOCS and HDRS scales; GHQ-5 was used to screen the normal controls. P300 was recorded by an auditory two-tone discrimination task in both groups using Neuropack Sigma-8 Nihon Kohden software. Results: P300 amplitude did not differ between patients and controls; neither was it correlated with most of the demographic and clinical variables. Patients had a significantly longer latency of P300 especially in frontal areas, those on medication also having a significantly longer latency than those off drugs. However, P300 latency did not vary significantly with sex, age, education, onset, duration, or course of illness, and severity of depression or obsessions & compulsions. Conclusion: This study replicated previous ones in showing no significant difference of P300 amplitude in OCD patients compared to controls. P300 latency was significantly increased (p = 0.035, 0.024 & 0.044, in Fz, F4 & F3, respectively) in frontal leads, suggesting an electrophysiological abnormality in this part of OCD brain, correlating well with other neuro-imaging and neuropsychological studies that have also implicated the frontal lobe in OCD.