Magnetoencephalographic characterization of sleep spindles in humans.

Sleep spindles in EEG recordings of adults are most prominent over the central and frontal midline regions. Early magnetoencephalographic recordings agreed with conventional EEG findings. However, more recent small-array magnetoencephalography and quantitative EEG studies suggest that the source areas for spindles are more widespread. We used a whole-head 122-channel biomagnetometer to characterize the sources of sleep spindles in four normal volunteers. Parallel interactive and automated multiple dipole spatiotemporal source modeling was conducted on the data sets of 10 spindles from each subject. Principal component analysis was used to estimate the number of sources in interactive source modeling, and singular value decomposition was used in automated dipole modeling. Spectral analysis of the epochs containing sleep spindles was performed. Principal component analysis and singular value decomposition suggested that all sleep spindles were made up of activity from multiple sources. Similarly, interactive and automated multiple dipole source modeling showed that three or more sources were present in 75% of spindle bursts. The sources for sleep spindles localized to all four cerebral lobes. Parietal and frontal lobes were the areas most frequently involved. Interactive source modeling resulted in more frequent temporal lobe than occipital dipole localizations; automated source modeling showed more frequent occipital than temporal sources. Spindle source localizations varied across subjects and across different spindles within subjects. Our results indicate that individual sleep spindles are generated by multiple cortical sources that are widespread within and across individuals.

[1]  F. Richer,et al.  Methodological considerations for the evaluation of spatio-temporal source models. , 1991, Electroencephalography and clinical neurophysiology.

[2]  M Huang,et al.  Multi-start downhill simplex method for spatio-temporal source localization in magnetoencephalography. , 1998, Electroencephalography and clinical neurophysiology.

[3]  J C Jones,et al.  The utility of placing sphenoidal electrodes under the foramen ovale with fluoroscopic guidance. , 1995, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[4]  H. Schulz,et al.  Topographical analysis of sleep spindle activity. , 1992, Neuropsychobiology.

[5]  N Nakasato,et al.  Magnetic detection of sleep spindles in normal subjects. , 1990, Electroencephalography and clinical neurophysiology.

[6]  D. Cohen,et al.  MEG versus EEG localization test using implanted sources in the human brain , 1990, Annals of neurology.

[7]  S. Supek,et al.  Simulation studies of multiple dipole neuromagnetic source localization: model order and limits of source resolution , 1993, IEEE Transactions on Biomedical Engineering.

[8]  J Hasan,et al.  Quantitative topographic electroencephalographic mapping during drowsiness and sleep onset. , 1995, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[9]  P. Achermann,et al.  Spindle frequency activity in the sleep EEG: individual differences and topographic distribution. , 1997, Electroencephalography and clinical neurophysiology.

[10]  M. Kajola,et al.  Generator sites of spontaneous MEG activity during sleep. , 1992, Electroencephalography and clinical neurophysiology.