The evoked K-complex: all-or-none phenomenon?

The functional significance and topographical variation of the different components of the evoked K-complex were examined. In the first experiment, the intensity of the stimulus (80 and 60 dB SPL) and its rise-and-fall time (2 and 20 milliseconds) were manipulated during nonrapid eye movement sleep. In the second experiment the tonal frequency (500, 1,000 and 2,000 Hz) of the stimulus was manipulated. In the first experiment, nine stimuli were presented every 10 seconds, whereas in the second, 20 consecutive stimuli were presented. The evoked K-complex consisted of two different negative components peaking at approximately 350 and 550 milliseconds, respectively, and followed by a positive component peaking at approximately 900 milliseconds. K-complexes were easier to elicit for high-intensity fast rise-and-fall time stimuli than for low-intensity slow rise-and-fall time stimuli. The probability of occurrence was not affected by the tonal frequency of the stimulus. When a K-complex was evoked, the amplitude and latency of N350, N550 and P900 remained invariant regardless of its intensity, rise-and-fall or its tonal frequency. The N550-P900 portion of the K-complex therefore appears to be an all-or-none phenomenon. On trials in which a K-complex could not be elicited, N350 was still visible although much attenuated. In these trials, its amplitude was further reduced when stimulus intensity was lowered. N350 might need to reach a certain critical threshold before the much larger N550-P900 complex is elicited.