Previous research indicated that duration could be a confounding factor in EEG research. Therefore the effect of stimulus duration on eventrelated potentials (ERPs) was assessed in this study, using faces as visual stimuli. The study consisted of a rapid serial visual presentation (RSVP) task with loguniformly distributed stimulus durations (1001500 ms) and no interstimulus intervals (ISIs), as well as an identical control task with varying ISIs (8002500 ms). Results showed no significant effect of stimulus duration on the P100, N170 and P300 ERP component peaks, but a significant positively correlated effect on the late stage of the P300 component (~300500 ms latency). Further the study revealed a significant effect of ISIs on the P100, N170 and P300 components peak amplitudes. During a brief additional analysis an adaptation effect regarding the N170 component during a sequence of visual stimuli (face images) was observed. Finally a qualitative assessment of the results indicated that a combination of deconvolutionbased overlap correction and stimulus duration modelling is methodologically sound.