Eye blinking artifacts present serious problems for electroencephalographic (EEG) interpretation and analysis. In this study, we apply independent component analysis (ICA) to eye blinking artifact removal from cognitive EEG recordings. Due to the specific design of the experiment, the eye blinks almost always co-occur with the event-related potentials (ERP), which creates problems for ICA. We introduced another data set of spontaneous blink and combined it with single-trial ERP data. Our results show that ICA on the combined data set gives separation that makes more sense and makes it easier for EEG interpretation and analysis