Evidence of a nonlinear human magnetic sense

Neuroscience. 2007 Jan 5;144(1):356-67. doi: 10.1016/j.neuroscience.2006.08.068. Epub 2006 Oct 25.


Human subjects respond to low-intensity electric and magnetic fields. If the ability to do so were a form of sensory transduction, one would expect that fields could trigger evoked potentials, as do other sensory stimuli. We tested this hypothesis by examining electroencephalograms from 17 subjects for the presence of evoked potentials caused by the onset and by the offset of 2 G, 60 Hz (a field strength comparable to that in the general environment). Both linear (time averaging) and nonlinear (recurrence analysis) methods of data analysis were employed to permit an assessment of the dynamical nature of the stimulus/response relationship. Using the method of recurrence analysis, magnetosensory evoked potentials (MEPs) in the signals from occipital derivations were found in 16 of the subjects (P<0.05 for each subject). The potentials occurred 109-454 ms after stimulus application, depending on the subject, and were triggered by onset of the field, offset of the field, or both. Using the method of time averaging, no MEPs were detected. MEPs in the signals from the central and parietal electrodes were found in most subjects using recurrence analysis, but no MEPs were detected using time averaging. The occurrence of MEPs in response to a weak magnetic field suggested the existence of a human magnetic sense. In distinction to the evoked potentials ordinarily studied, MEPs were nonlinearly related to the stimulus as evidenced by the need to employ a nonlinear method to detect the responses.

Publication types

  • Clinical Trial

MeSH terms

  • Adolescent
  • Adult
  • Electroencephalography
  • Electromagnetic Fields*
  • Evoked Potentials / physiology
  • Female
  • Humans
  • Linear Models
  • Male
  • Middle Aged
  • Nonlinear Dynamics
  • Sensation / physiology*