"When men wish to construct or support a theory, how they torture facts into their service!"
Even in 1852, psychologists like Charles Mackay, who wrote those words in his book Extraordinary Popular Delusions and the Madness of Crowds, were well aware of the dangers of confirmation bias.
I was reminded of the pervasiveness of this cognitive bias last weekend during a visit to GlaxoSmithKline's vaccine distribution facility in Marietta, Pennsylvania. Confirmation bias is as dangerous in 2010 as it was in 1852.
I was invited by David Wescott to join a group of bloggers for an event at GSK's facility during which they would discuss the importance of adult vaccination, and I agreed to participate. My travel, accommodations and food were covered by GlaxoSmithKline. There was no explicit or implicit expectation that I'd write about the trip, and they have not reviewed this post. (Quite an impressive release of control for a multinational pharmaceutical corporation!)
As a science blogger, it wasn't exactly clear why I was included, as science bloggers generally already "get it." We know that vaccination is important, and we often write about it. That public health is a bit beyond the scope of this particular science blog only further served to confuse me. Then I landed in Philadelphia and arrived at the hotel to meet up with David and the other invited bloggers. I found myself surrounded by a relatively diverse group of intelligent bloggers...but, well, they were all mom bloggers, and all relatively local. Why did they fly me all the way out from LA? Why was I there? I still couldn't figure it out.
The next morning, after the two hour ride to Marietta (during which I observed that the changing of the leaves during autumn is NOT a myth!) the program started. It was then that I finally understood why I was there.
I don't write about health, or vaccines, I don't have a family, and I already understand the importance of vaccination. But I do write about the mind. And while all the women were watching the presentations on adult vaccination, I was watching them. At one point, the speakers were discussing the importance of getting inoculated each year against the influenza virus. I don't mean to single anyone out, but one comment was particularly striking: "I've never gotten the flu vaccine, and I've never gotten the flu." As if to suggest that there is no real reason she should have to worry about the flu virus. The very same flu virus that evolves resistance to our vaccinations so rapidly that new vaccine formulas need to be developed each year. The same flu that kills thousands in this country each year. "How they torture facts into their service," indeed.
The thing is, though, that this sentiment, which betrays a certain misunderstanding of epidemiology, statistics, and biology, was not uttered by an anti-vaccination wingnut praying at the altar of Jenny McCarthy. The woman vaccinates her children! She gets it! She is an unfortunate victim - not of anti-vaccination propaganda or the so-called "balanced" journalism - but of the cognitive biases built into her own mind.
Consider a group of people in a coin-flipping tournament. If someone wins a coin flip, they stay in the game. If they lose, they're out. At the end of fifty rounds of coin-flipping, there will be one person left, who will have won the tournament. She will have won the coin-flip fifty times in a row. But there is nothing special about her. When it comes to the fifty-first coin flip, she still has an equal probability of winning or losing. But our minds aren't really equipped to understand this kind of probability.
That brings me to my second observation. For a corporation that depends on communicating science to the public, they did a terrible job of it! For a certified card-carrying data-whore like myself, the Powerpoint presentations (which broke every. single. rule. of effective presentations. I highly recommend that they hire my friend Les Posen to teach them how to present properly) did not have enough detail. They would present some statistic, but without the level of detail required for me to make any real sense of the data. I will grant that I was not the intended audience of the talk, so I will forgive them their lack of error bars and missing p-values. For a general audience without a scientific data-driven background, the presentation was even more useless! It was all statistics, bar graphs, and numbers. If you're going to communicate science to a general audience (and I'd like to think that I know a thing or two about communicating science to a general audience), you need to engage them emotionally. You need to tell a story, not drown them in statistics. The presenter might say something like, "Last year, three gazillion people died because they were not vaccinated against a Terrible Disease That Kills People In Gruesome Ways But Which We Could Eradicate In Less Than A Decade If Everyone Would Just Get Vaccinated." Everyone would agree that this is a Bad Thing, at least. More likely, it is a Terrible, Horrible, No Good, Very Bad Thing.
But the human mind is not capable of really understanding the magnitude of a gazillion. To us, a gazillion and fifty thousand and twenty thousand aren't really so different. For that matter, neither are twelve and sixteen. That is because as the magnitude of a number increases, the specificity of our mental representation of that number decreases. Put another way, as the magnitude of a number increases, the margin of error in forming an accurate mental representation of that number increases as well. This is also true for other animals, like monkeys and rats and fish.
Giving us incidence and death rates and other such statistics doesn't really get the job done. It doesn't communicate what they want it to. Nor will glossy pamphlets (like the one they gave me) featuring Mia Hamm telling us to get vaccinated. What will get the job done is story-telling, appealing to emotion, and utilizing accessible analogies. Instead of telling us how many gazillions died last year, tell us how many airplanes full of people, or how many football stadiums full of people died last year.
The people who work on developing vaccines put years and years of effort into developing their products. And corporations like GSK spend millions of dollars funding those efforts. It is a shame that they can't communicate the critical importance of vaccination in a more effective way.
As we left the facility, I turned to David and quietly asked him, "do companies like this have any psychologists in their communications departments?" "No," he said, "they're all PR and marketing people."
Perhaps they should consider hiring on a cognitive psychologist or two.
Addendum: It occurs to me that GSK's legal department must carefully scrutinize any sort of public communication, and perhaps the presenters we interacted with were severely restricted in terms of content and delivery style. That said, there must be a way to communicate science in a more robust way that still falls within the confines of legal acceptability. An interesting question to ponder.
Here's the disclaimer again: My travel, accommodations and food were covered by GlaxoSmithKline. There was no explicit or implicit expectation that I'd write about the trip, and they have not reviewed this post.
Learn more about vaccines, courtesy of our friends at Science-Based Medicine.
Posts from some of the other bloggers at the event:
Coverage from the official GSK blog: More Than Medicine.