This review summarises all the evidences about the influence of different vigilance states on the occurrence of spike wave discharge (SWD) in idiopathic generalised epilepsy (IGE) patients. Numerous converging observations showed that full REM-sleep and alert wakefulness exert strong inhibition. A critical zone of vigilance which is a transitional state between waking and non-REM (NREM) sleep, and NREM sleep and REM sleep, has a promoting effect on the absence type spike wave discharge. Spike wave discharges are associated with phasic arousals without awakening and are attached to oscillation son the microstructural level of sleep, perpetuated by cyclic arousal events known as 'cyclic alternating pattern' (CAP), especially within the critical zone, but also along the whole sleep process. More specifically SWD seems to be attached to the 'A-phase' of CAP which is a reactive one and reflects synchronised NREM sleep EEG elements, like K-complexes, spindles and delta groups. The more slow wave elements are found in phase A--like in subtype A1--the more the coincidence with SWD occurs, and the more it is characterised by fast rhythms--as in subtype A2 and A3--the less the association with SWD could be observed. Since subtype A1 is associated with the first sleep cycle and with the descending branches of cycles, it is concluded that SWD appear in those dynamic moments of vigilance level oscillations which were characterised by strong sleep-like answers to arousal influences in high sleep pressure periods of sleep cyclicity. These data harmonize with another line of evidence suggesting that SWD represent the epileptic variant of the complex thalamocortical system function which is the substrate of NREM sleep EEG phenomena. In idiopathic generalised epilepsy there is a growing body of evidence that--as it was assumed by Gloor--spindles transform to SWD pattern. These data explain why those dynamic changes which evoke sleep responses are promoting for the occurrence of SWD. Adapting these data we offer a new interpretation to explain the strong activation effect of sleep deprivation in this kind of epilepsy. We assume that it is mainly due to the forced vigilance level oscillations, especially in morning, when elevated sleep pressure and circadian wake promoting forces, representing opposite tendencies, increase the amount of oscillations.