Transient decoupling of cortical EEGs following arousals during NREM sleep in middle-aged and elderly women

Int J Psychophysiol. 2010 Aug;77(2):71-82. doi: 10.1016/j.ijpsycho.2010.04.007. Epub 2010 May 5.


Spontaneous cortical arousals in non-REM sleep increase with age and contribute to sleep fragmentation in the elderly. EEG spectral power in the faster frequencies exhibits well-described shifts during arousals. On the other hand, EEG activities also exhibit correlations, which are interpreted as an index of interdependence between distant cortical neural activities. The possibility of changes to the interdependence between cortical regions due to an arousal has not been considered. In this work, using previously recorded C3A2 and C4A1 EEG signals from two groups of adults, middle-aged (42-50 years) and elderly (71-86 years) women, we examined the effects of spontaneous arousals in NREM sleep on cortical interdependence. We quantified the auto- and cross-correlations in these signals using mutual information and characterized these correlations in periods before the onset and following the end of arousals. The pre-arousal period exhibited significantly higher interdependence between central regions than that following the arousal in both age groups (middle-aged: p=0.004, elderly: p<0.0001). Also, for both EEG signals the auto mutual information had a faster rate of decay, implying lower signal predictability, following the arousal than prior to it (both age groups, p<0.0001). These results indicate that the state of the cortex is different after, compared to before, the arousal even when the spectral power changes characteristic of an arousal are no longer visible. The findings suggest that the state following an arousal characterized by lower interdependence may resemble a more vigilant period during which the system may be vulnerable to more arousals.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Age Factors
  • Aged
  • Aged, 80 and over
  • Arousal / physiology*
  • Cerebral Cortex / physiology*
  • Electroencephalography* / methods
  • Female
  • Humans
  • Middle Aged
  • Polysomnography / methods
  • Sleep Stages / physiology*
  • Time Factors