Recently, readers have e-mailed to inform me that I am a thug-hugging, dope-smoking member of the liberal media elite that is cramming its stinking socialist agenda down the throats of ordinary Canadians. Oh, and I hate Western civilization.

Other readers do not share that opinion. They have been in touch to explain that I am a climate-change-denying, cancer-promoting stooge of the corporate media that is cramming its neo-conservative agenda down the throats of ordinary Canadians. And I hate David Suzuki.

Occasionally, these e-mails arrive one after another in my inbox and I am left with the impression that I am a dope-smoking, neo-con socialist cramming David Suzuki's corporate agenda down the throats of ordinary Canadians. Which is very confusing.

This is standard stuff for journalists, and not only opinion writers. People with strong views react strongly to what they see in the media. They also react predictably.

Psychologists call it the "hostile media phenomenon": Someone with a strong opinion about a politician or an issue is far more likely to perceive bias in the media's coverage of that politician or issue than is someone who does not feel so strongly. And the bias is almost always seen to be against the position of the person who perceives it. Hence, the media are "hostile."

Research on the hostile media phenomenon began with a group of psychologists at Stanford University. In 1980, three days before the presidential election that pitted Jimmy Carter against Ronald Reagan, the researchers surveyed 160 registered voters and asked them whether the media had covered both candidates fairly or whether coverage had been biased one way or another.

Two-thirds said the media had been fair. One-third said it was biased. Of those who said the coverage was slanted, 89 per cent said the bias was against the candidate they supported.

This is unsurprising -- to use a gross understatement -- to anyone who has worked in a newsroom. We get this all the time, but it's particularly common during elections.

Strong supporters of one party complain the media is wildly, blatantly, shamefully biased against that party. Partisans on the other side are equally furious because they see precisely the opposite. To both, the media's bias is so obvious only fools and liars would deny it.

To dig deeper into this phenomenon, the Stanford psychologists created an experiment shortly after the 1982 massacre in the Lebanese refugee camps of Sabra and Chatila. The massacre was a huge story at the time, in part because Israel's responsibility was hotly disputed.

Using selective recruitment and surveys, the researchers formed three groups of Stanford students. The first was strongly pro-Israeli. The second was pro-Arab. The third was made up of students who didn't have strong feelings one way or the other.

All the students were shown six segments from television news broadcasts about the massacre and the controversy. They were then asked to answer a detailed questionnaire.

The results were what any journalist would expect them to be. Pro-Israeli students said the reports were strongly biased against Israel. Pro-Arab students said they were strongly biased against Arabs. And neutral students fell pretty much in between (they collectively felt the reports were very slightly anti-Israel).

The pro-Israeli and pro-Arab students also tended to be confident that they knew the personal views of the journalists who put the news together. "Pro-Arab students believed that the editorial staffs were somewhat pro-Israel," the psychologists wrote, "whereas the pro-Israeli students believed they were somewhat anti-Israel."

When the researchers looked more carefully at the results within the partisan groups, they found that the more passionately someone felt, the stronger was their sense that the media was biased against their side. The researchers also found that students who felt they knew a lot about the situation gave higher ratings of media bias than those who said they knew less.

This, along with later research, demonstrated the phenomenon exists on a sliding scale: It is most extreme in fiercely committed partisans who believe they are well-informed on the issues, but it also shows up in more moderate form among those with more moderate views.

Lots of other studies have confirmed these findings but probably the most intriguing is a 2004 experiment by researchers Albert Gunther and Kathleen Schmitt. They did something very neat and simple with a group that was equally composed of people with strong feelings against genetically modified organisms and people who supported GMOs.

Half the test subjects were asked to read a piece of writing about GMOs. It's a newspaper article, they were told. The other half were given the same writing but they were told it was actually a student's essay.

The group that thought they were reading a newspaper article reacted predictably: Those who were pro-GMO thought the article was biased against GMOs, while those were anti-GMO were sure it was biased in favour of GMOs.

Meanwhile, the group that was told the writing came from a student paper detected no hostile bias. In fact, both pro- and anti-GMO people tended to believe the article supported their view.

What psychological mechanisms are at work here? Researchers have debated that for years but there's still no conclusive answer.

The only thing we know for sure is that this phenomenon is real. And it's a good reason for each of us to carefully examine our own conclusions for bias before we get angry and send a letter to the editor to complain about the media's bias.

Of course it is also possible that this whole column is nothing more than a ruse to deflect attention from the masters I serve and the agenda we are cramming down the throats of ordinary Canadians. Readers who wish to condemn your correspondent, along with said masters and agenda, are directed to the e-mail below.