Can People Sense a 60 Hz Magnetic Field?
A slightly edited excerpt from Are Electromagnetic Fields Making Me Ill? by Brad Roth
Robert Tucker and Otto Schmitt, working in the Electrical Engineering Department at the University of Minnesota in the 1970s, decided to test if people could sense a 1-millitesla, 60-hertz magnetic field (IEEE Transactions on Biomedical Engineering, Volume 25, Pages 509–518). Such a field is two hundred times greater than is typical from a power line, and twenty times stronger than the earth’s static field. Their initial experiment revealed that some “super perceivers” were able to tell reliably if a magnetic field was present, with a p-value of 0.000000000000000000000000000001. The p-value corresponds to the probability that a result can arise by chance; generally a p-value less than 0.05 is considered “statistically significant.” There was no doubt these results were significant, and that subjects could accurately predict when the field was on.
Some researchers might have published this result, claiming that they had discovered magnetic perception in humans. Tucker and Schmitt were skeptical, wondering if these super perceivers might not be sensing the magnetic field, but instead be detecting non-magnetic clues. So they launched a five-year effort to rid their experiment of any hint that the field was turned on.
Tucker and Schmitt’s article is fascinating, as they recount this effort in detail. They replaced the human experimenter by a computer that determined randomly if the magnetic field would be on or off during each test. At a location too far away to interfere with their experiment they built a second electromagnet identical to the real one used to apply the magnetic field. With it, they could avoid any clues from the slight dimming of the building lights caused by turning on the large current through the coil, or any soft hum associated with movements of the coil, or the clatter linked to turning relays on and off. They built an “acoustically padded cabinet” to further isolate the subject from sound and vibration. These efforts increased the p-value to 0.00000000000000000001, which still meant detection was highly significant.
Tucker and Schmitt redoubled their effort. They built a new plywood cabinet to house the experiment, weighing over 600 pounds. They constructed it using no iron nails or hinges that could interact with a magnetic field, but instead fastened the cabinet together with a few brass screws. The door was seated on a rubber gasket and sealed by pulling a slight vacuum. Their next rounds of tests, carried out in this cabinet, showed that a few people could still detect when the magnetic field was on, with a p-value up to a still-significant 0.0000000001.
Scientists must be persistent. Tucker and Schmitt proved their tenacity with their next modifications. They hung the cabinet from the roof by an aircraft bungee shock cord to isolate it from even the mildest vibrations. To prevent the cabinet from swinging, they placed four slightly inflated automotive inner tubes between the floor and the cabinet base. They reclamped the turns of the coil to prevent any intracoil “buzzing.” They draped the inside of the cabinet with sound absorbing material, and they shock mounted the subject’s chair. After this Herculean struggle, they performed their final experiment, and found that no subjects could detect a 60-Hz magnetic field. Their article concluded:
“A large number of individuals were tested in this isolation system with computer randomized sequences of 150 trials to determine whether they could detect when they were, and when they were not, in a moderate (7.5–15 gauss [0.75–1.5 millitesla] rms [root mean square]) alternating magnetic field… In a total of over 30,000 trials on more than 200 persons, no significantly perceptive individuals were found, and the group performance was compatible, at the 0.5 probability level [a 50/50 chance expected for random guessing], with the hypothesis that no real perception occurred.”
The Tucker and Schmitt experiment teaches a vital lesson. Scientific experiments are difficult, and scientists must work doggedly to eliminate any subtle systematic errors. As experiments improve, effects apparent in preliminary studies may become smaller and smaller, until disappearing once all sources of error are gone. Removal of artifacts is especially important when trying to distinguish a weak response from no response at all. One reason the bioelectric literature is filled with inconsistent results may be that not all experimenters are as diligent as Robert Tucker and Otto Schmitt.