Expectation bias cuts both ways, for positive and negative expectations. Expectation bias, the tendency to perceive and accept data that reinforces your expectation, is one of the many contributors to placebo effects (the illusion of a positive benefit that derive from something other than an active treatment). It is also, however, part of nocebo effects  (the illusion of negative side effects from something other than active treatment).

Expectation bias can be powerful enough in some people to lead not only to the perception of a benefit or side effect but to a frank delusion. When this happens on a large scale, that can lead to a mass delusion. There are many episode that demonstrate this effect, but now there is also a controlled experiment that also confirms it.

A recent study looked at sham exposure to wifi signals in 147 subjects. They were first exposed to either a documentary about the dangers of wifi, or to a documentary about internet security. A total of 54% of the subjects experienced

“…agitation and anxiety, loss of concentration or tingling in their fingers, arms, legs, and feet. Two participants left the study prematurely because their symptoms were so severe that they no longer wanted to be exposed to the assumed radiation.”

Further, the group exposed to the wifi documentary experience significantly more symptoms.  This is a small study but it matches prior research showing that those who believe they have electromagnetic sensitivity will experience symptoms when exposed to sham EMF. The difference with the current study is that it used healthy volunteers and controlled for media exposure.

Systematic reviews of the research on EM hypersensitivity show that those who self-identify as having EM hypersensitivity (which has now been renamed in the technical literature as “Idiopathic environmental intolerance attributed to electromagnetic fields”) cannot tell the difference between real and sham EMF. This review concluded:

” No robust evidence could be found to support this theory. However, the studies included in the review did support the role of the nocebo effect in triggering acute symptoms in IEI-EMF sufferers. Despite the conviction of IEI-EMF sufferers that their symptoms are triggered by exposure to electromagnetic fields, repeated experiments have been unable to replicate this phenomenon under controlled conditions.”

The new study suggest that this nocebo effect can happen on a large scale due to media reports, and cautions the media about sensationalizing such reports.

Perhaps the most famous recent example of this phenomenon is the Pokemon Contagion episode. In December 1997 it was reported widely by the media in Japan that one  particular episode of Pokemon that involved flashing lights provoked seizures in a few susceptible children (they had photosensitive epilepsy). This led to over 12,000 reports of symptoms among children watching the episode. Analysis of cases found that only a very few cases were legitimate seizures, the rest appeared to be anxiety provoked by the media reporting.

Robert Bartholomew is a sociologist who specializes in delusions. He wrote an excellent article for the Skeptical Inquirer in which he documents many of the mass delusions over the last millennium. My favorite is the Seattle Windshield Pitting Episode of 1954. The observation of small pits in the windshield of a car led many in the public to examine their own windshields, finding similar pitting. This led to wild speculation about atomic fallout and other causes.

What is clear from both research and historical cases is that media reporting can provoke a self-fulfilling phenomenon by creating the very symptoms that are being reported. Ideally, media outlets will show restraint and professionalism when reporting on possible medical risks. (I’ll stop there as your hysterical laughter will probably keep you from being able to read further.)