Power spectrum analysis and heart rate variability in Stage 4 and REM sleep: evidence for state-specific changes in autonomic dominance.

The present study investigated autonomic activity during NREM and REM sleep stages and wakefulness by spectral analysis of heart rate variability. The results demonstrated that NREM sleep in humans was characterized by a widely different autonomic activation pattern than REM sleep: high parasympathetic activity was found in NREM, while REM was characterized by attenuated vagal tone, and augmented sympathetic activity. The overall pattern during wakefulness showed an intermediate position between NREM and REM patterns; parasympathetic activity was lower than in NREM and higher than in REM, with an opposite trend for sympathetic activity.

'Oddball' event-related potentials and information processing during REM and non-REM sleep.

Auditory stimuli consisting of the subject's own name and an irrelevant word, counterbalanced in probabilities, were presented to 15 male subjects in the awake state and during natural sleep. Potentials recorded to these stimuli, as well as to clicks presented during sleep in a preceding night, were recorded and compared. Principal component analysis (PCA) was conducted on evoked potentials to distinguish temporally overlapping components, and ANOVA was applied on the eigenvector coefficients. During non-REM sleep a parietal P450, more prominent in stage 2, was observed in addition to the prominent waveform of a K-complex, which was also recorded in response to clicks and consisted of N350, N550 and P1000. During REM sleep, a fronto-central negativity which resembled non-REM N350, a parietal positivity at about 450 ms and a large N700 were detected. ANOVA on PCA coefficients showed a significant effect of verbal-stimulus type (name/irrelevant) on an eigenvector, which included all the components observed during stage 2 sleep. Coefficients during REM sleep showed a significant effect of stimulus probability on an eigenvector consisting of a prominent P450, suggesting a resemblance to the awake P300 component. This could not be demonstrated during non-REM sleep. The results indicate continued evaluation of auditory input salience during sleep, which diminishes during deep sleep, and is replaced by evaluation of stimulus context in a train of stimuli during REM sleep.

P300 in response to the subject's own name.

The P300 component is influenced by task relevance which is typically achieved by employing an active task involving the evoking stimulus. This study explored the possibility of making use of the subject's name, which is innately relevant to the subject, to achieve stimulus relevance. Dependence of P300 amplitude on auditory presentation of the subject's name was assessed in two experiments in which no response was required: (1) a passive 2-word "oddball" paradigm where the low probability word was the subject's first name; (2) a passive 3-word "oddball" paradigm consisting of two low probability words, one of which was the subject's name, and a third, high probability word. In the first experiment, a typical active 2-word "oddball" paradigm, which did not include the subject's name, was compared with the passive paradigm. P300 amplitude was larger in response to the subject's name compared to the other word in the two-word paradigm, indicating that the name can be used to evoke P300 in a passive paradigm. It was also larger than either of the other two words in the 3-word paradigm, suggesting that stimulus relevance has an additional effect on P300 amplitude beyond rarity.