July 2011 Archives

Diabetes and obesity in America.pngI recently came across an article in the online magazine, Slate, which reminded me of the importance of using the principles of scientific thinking. The article describes a study that got lots of media attention--including an appearance for one of the authors on Stephen Colbert's TV show-- because its researchers claimed to have found that many behaviors--obesity, divorce--spread "like a virus" through social networks.

This claim that behavior is contagious is extraordinary, and, according to our textbook (page 22), "extraordinary claims require extraordinary evidence." In addition, not only was the claim extraordinary, people who looked critically at the claim found that the study had not passed peer-review or been published in a scientific journal; the researchers (it turned out) were using some questionable methodology. All the media attention came from a paper posted to an online website for papers in progress. So the scientific principle of "replicability" was also not met because, according to the Slate article, the results have not been replicated in other studies using the researchers unorthodox methodology .

One of the reasons that the study was trusted, according to the original article, was that the authors are respected scientists who have published in some of the most prestigious medical journals in the world. I think a second reason would be "confirmation bias" (Lilienfeld, page 8). Even the article in Slate falls prey to this bias. It concludes, "One irony of the contagion battles is that even if their methods are suspect, Christakis and Fowler are obviously correct that peer influence exists and that it may be even more important than we realize."  The author of the Slate article provides no evidence to support this assertion, he is just asserting something that many people want to believe.

I think there is a third reason why the study was trusted, which seems to be related to confirmation bias. Right now, the media are obsessed with social media like Facebook and Twitter. In addition, obesity in the USA is a problem with huge consequences and costs, and people are desperate to figure out why. This was a novel, trendy explanation for a significant social problem, so people wanted to believe it. 

 I've posted a chart from a Harvard university medical journal that illustrates the obesity epidemic and its consequences. You can see the change in the prevalence of obesity (left column) and diabetes (right columna) between 1994 and 2005. We know that type II diabetes is caused by obesity; its rising prevalence is associated with rising health care costs. So American journalists were eager to embrace a study that explained a serious issue (obesity) in terms of a popular phenomenon (social contagion) and ignored the evidence that this was an extraordinary claim and not replicated. (453 words)

Dalmatian illusion.pngHere's a classic visual illusion of a dalmatian dog in a sun-dappled field, but really, all you see are black dots against a white background. Our brain makes sense out of this black and white stimuli, and "sees" the dog.

This summer I was hiking and suddenly, from the underbrush, in romped the dog seen in the second photo. He posed briefly, then dashed off to join his walking companion.

Dalmatian in France.pngAnd thus does life mimic art.

About this Archive

This page is an archive of entries from July 2011 listed from newest to oldest.

September 2011 is the next archive.

Find recent content on the main index or look in the archives to find all content.