The shape of evoked potentials is influenced by the level of vigilance, varying with sleep–wake states. In this paper the shape of auditory evoked potentials is modelled by taking two factors, both modulating the underlying neuronal substrate, into account: ‘sensory gating’ and ‘neuronal firing mode’. Under low levels of vigilance sensory gating reduces the amount of neuronal activity reaching the cortical centres. Due to a rise in hyperpolarisations of thalamocortical neurons associated with an increasing depth of sleep, stimulus evoked primary and secondary excitations, seen as correlates of the N1 and N2 waves of the evoked potential, become smaller. Heightened hyperpolarisations also change the spontaneous activity of neurons from the ‘tonic’ firing mode of wakefulness into the ‘burst–pause’ firing mode of sleep. The large P220 complex together with the N350 and N550 waves in sleep are caused by the stimulus induced triggering of pauses and bursting of neurons. The results of this modelling experiment confirm the view that sleep-specific components such as P220, N350 and N550 are waves that facilitate and protect sleep, whereas the wake-specific components N1, P2–P3 and N2 have perceptual–cognitive functions. In particular the wake P2–P3 wave is sensitive to cognitive functions, such as attention. Based on the modelling results it is suggested that component negativities, expressed in N1, N2 and N350, reflect excitatory processes, whereas positivity in P2–P3 and P220 is a correlate of inhibitory processes. Hence, the large P3 in an attended condition is also interpreted as an inhibitory process suppressing irrelevant information, facilitation the saliency of relevant information.
Evoked potentials in the electroencephalogram (EEG) form a main research tool to study processes of vigilance during sleeping and waking, as well as the amount of information processing during these states. Shape and amplitude of these evoked potentials are considerably modulated by vigilance processes, indicating major variations in the processing of external information during sleep–wake states. Since evoked potentials are independent of behavioural responses and conscious awareness, they form an adequate tool to establish the extent of information processing during all states of vigilance. In this framework evoked potentials are frequently studied during sleep and wake states. Recent reviews of vigilance modulation of human evoked potentials are presented by Bastien et al., 2002, Campbell and Colrain, 2002, Colrain and Campbell, 2007 and Yang and Wu, 2007. N1 is mainly controlled by the physical features of the stimulus, as well as by the general state of the subject's brain. N1 is largest under alert wake conditions and decreases at lower levels of vigilance (Nordby et al., 1996, Campbell and Colrain, 2002 and Campbell, 2010). N1 is regarded as indicative of stimulus registration and detection, presumably underlying the perception of the stimulus (Näätänen and Picton, 1987, Campbell and Colrain, 2002 and Campbell, 2010). The P2 component (or P220) on the other hand shows an amplitude increase during sleep (Nordby et al., 1996). The amplitude is even larger in deep sleep than in light slow wave sleep. The P220 wave is a prominent sleep component of the evoked potential (Bastien et al., 2002), but its significance and function is still not fully clear (Crowley and Colrain, 2004). After this positive sleep wave a negative wave follows, in the time domain of 300 to 400 ms. and mostly indicated as N350. When subjects further descend into sleep later negativity start to increase getting its maximal amplitude between 500 and 750 ms. This negativity is also known as N550 or the ‘late negative wave’. The long latency of this component seems to indicate that processing of external sensory stimuli is delayed (Nordby et al., 1996). The shape of the evoked potentials constructed during REM sleep is mostly fairly similar to those of waking, although the amplitude of N1 is smaller compared to its waking level (Muller-Gass and Campbell, 2010).