Human cortical oscillations: a neuromagnetic view through the skull

https://doi.org/10.1016/S0166-2236(96)10065-5Get rights and content

The mammalian cerebral cortex generates a variety of rhythmic oscillations, detectable directly from the cortex or the scalp. Recent non-invasive recordings from intact humans, by means of neuromagnetometers with large sensor arrays, have shown that several regions of the healthy human cortex have their own intrinsic rhythms, typically 8–40 Hz in frequency, with modality- and frequency-specific reactivity. The conventional hypotheses about the functional significance of brain rhythms extend from epiphenomena to perceptual binding and object segmentation. Recent data indicate that some cortical rhythms can be related to periodic activity of peripheral sensor and effector organs.

Section snippets

Source modelling

The invasive microelectrode recordings have the advantage that the neuronal structure generating the signals can be identified easily. By contrast, MEG (and EEG) sensors pick up signals from extensive brain regions, which might be even several centimetres away from the sensor. Therefore the sites of active neuronal populations have to be deduced from the measured signal distribution. Although this ‘inverse problem’ does not have a unique solution in the general case6,9, modelling the generators

Sources and reactivity of the alpha rhythm

Sources of the posterior 10 Hz alpha rhythm concentrate predominantly in the parieto–occipital region and, to a smaller extent, in the occipital areas. The two subclusters differ both by their sites and current orientations, and suggest that the strongest activity occurs in the parieto–occipital sulcus with less activity in the calcarine sulcus22,26. Even within these subclusters, several single sources with independent time behaviour might be active simultaneously11, thereby supporting

Segregation of the 10 Hz and 20 Hz somatomotor rhythms

The well-known ‘comb-like’ shape of the somatomotor mu rhythm implies that the rhythm consists of two or three frequency components with a nearly harmonic relationship. The dominant magnetic components near 10 Hz and 20 Hz (cf. Fig. 1) can occur either separately or simultaneously14, suggesting that the 20 Hz activity can be phase-locked to the 10 Hz rhythm at certain moments (and thus probably arise from the same neural generator) but stay totally independent at some other moments of time.

Hemispheric balance of somatomotor reactivity

Cortical signals associated with unilateral movements imply bilateral involvement of the somatomotor cortex. The bilaterality is particularly evident in the modulation of cortical rhythms, illustrated in Fig. 5 for unimanual finger movements46. Movement-related changes in the level of spontaneous activity occurred in both hemispheres but were strongest contralaterally, particularly at frequencies above 14 Hz. The 10 Hz activity had already started to dampen 2 s before the movement. In the

Functional significance of cortical rhythms

The available hypotheses for the role of cortical macroscopic oscillations include epiphenomena, with no functional significance, and idling, which would allow the system to start more rapidly than by cold start47. In agreement with the latter interpretation, the parieto–occipital and rolandic rhythms are strongest when the modality-specific sensory input (or motor output) is minimal or monotonous. In the visual system this happens when the eyes are closed or when one looks at a homogeneous

Concluding remarks

A surprisingly small number of synchronized cells might determine the recorded macroscopic signal that reflects the gross activity of a cell population26. Macroscopic cortical rhythms therefore probably arise in areas with the most synchronous signal transfer. It is worth emphasizing that changes in the synchronization of a neuronal population can occur without significant changes in the mean neuronal firing rates, and thus without changes of blood flow and metabolism. This means that

Acknowledgements

This work was supported by the Academy of Finland and by the Sigrid Jusélius Foundation. The magnetic resonance images were obtained at the Dept of Radiology, University of Helsinki. We thank N. Forss, J. Mäkelä, and S. Salenius for comments on the manuscript.

References (59)

  • M. Steriade

    Electroencephalogr Clin Neurophysiol

    (1990)
  • R. Salmelin et al.

    Electroencephalogr Clin Neurophysiol

    (1994)
  • R.M. Chapman

    Electroencephalogr Clin Neurophysiol

    (1984)
  • J. Tiihonen et al.

    Neuroscience

    (1989)
  • G. Chatrian et al.

    Electroencephalogr Clin Neurophysiol

    (1959)
  • J. Tiihonen

    Neurosci Lett

    (1991)
  • S. Salenius

    Electroencephalogr Clin Neurophysiol

    (1995)
  • B. Lütkenhöner

    Electroencephalogr Clin Neurophysiol

    (1992)
  • C. Tesche et al.

    Electroencephalogr Clin Neurophysiol

    (1993)
  • C. Perez-Borja

    Electroencephalogr Clin Neurophysiol

    (1962)
  • F. Lopes da Silva et al.

    Neurosci Lett

    (1977)
  • R. Salmelin et al.

    Neuroscience

    (1994)
  • P. Fletcher

    NeuroImage

    (1995)
  • G. Pfurtscheller

    Electroencephalogr Clin Neurophysiol

    (1981)
  • R. Salmelin

    NeuroImage

    (1995)
  • W. Kuhlman

    Electroencephalogr Clin Neurophysiol

    (1978)
  • S. Salenius

    Neurosci Lett

    (1996)
  • R. Llinás

    Science

    (1988)
  • M. Steriade et al.

    Physiol Rev

    (1988)
  • P. Andersen et al.

    Physiological Basis of the Alpha Rhythm

    Appleton–Century–Crofts

    (1968)
  • D. Cohen

    Science

    (1972)
  • S.J. Williamson et al.

    J Magn Magn Mat

    (1981)
  • R. Hari et al.

    Science

    (1989)
  • R. Hari
  • M. Hämäläinen

    Rev Mod Physics

    (1993)
  • A.I. Ahonen

    Physica Scripta T49

    (1993)
  • V. Vvedensky et al.

    Med Biol Eng Comput

    (1986)
  • H. Berger

    Arch Psychiatr

    (1929)
  • E.D. Adrian et al.

    Brain

    (1934)
  • Cited by (563)

    View all citing articles on Scopus
    View full text