The 'gap effect' (faster saccades to a peripheral target when its onset is preceded by the offset of the central stimulus) has often been referred as an example how attentional processes and oculomotor control interact. The neural processes underlying the gap effect were studied in a series of experiments by measuring high-density event-related potentials (ERP's) time-locked both to the target presentation and the saccade execution. We observed three neural correlates of the gap effect: (i) a frontal positivity / temporal negativity complex in response to the fixation offset or to other warning cues which may reflect specific preparatory processes, (ii) an enhancement of the early cortical visual responses to the peripheral target following either an offset or a change of the fixation stimulus similarly to the effect of attentional manipulations, and (iii) a prolongation of parietal activity before saccade execution when the fixation stimulus remained on which might be a correlate of a disengagement process. These results suggest that several factors contribute to the gap effect, each having its own neural basis, and that the cortical processes uncovered by these studies cannot explain the gap effect entirely. The results of similar studies in infants suggest that the behaviourally similar gap effect might be based on the functioning of different neuronal structures and that the cortical control of saccades undergoes a rapid development during the first year of life.