The spatiotemporal analysis of brain activation during the execution of same-different judgments [including the different color stimulus pair (DC), the different shape stimulus pair (DS) and the same stimulus pair (Same)] was performed in 12 normal young adult subjects using high-density event-related brain potentials (ERPs). We want to reveal the neural basis of “same” and “different” judgments. We can judge which is correct between the dual-process model and single-process model that previous experiments want to prove with behavioral research. The behavioral data showed that the same-different judgments yielded a robust fast-same effect as indexed by longer RT for DC and DS than Same. Scalp ERP analysis revealed the neurophysiological substrate of the fast-same effect: a more negative N2 component in DC and DS as compared to Same was found between 280 and 320 ms post-stimulus over midline fronto-central scalp regions. Following that period, DC and DS elicited a more negative P3 component than Same did about 450 ms after onset of the stimuli which might reflect the comparison and recognition of the partial character of the stimulus (the slow processing machine), and the latency of P3 was much shorter for Same than for DC and DS which was consistent with the behavioral data. Dipole source analysis (BESA software) of the original waveforms of DC and DS indicated that a generator of N2 mainly localized in the superior/anterior frontal cortex contributed to this effect, possibly in relation to early monitoring mismatch stimulus pairs. However, activation of the precuneus possibly resulted in N2 effect under Same condition, which was relative to holistic comparison (identify reporter). The results indicated that there might be different neural basis of “same” and “different” judgments. Our data support the dual-process model, and we think that IR and the slow processing machine work in order.