I'm Biased and So Are You

A month or so ago it was the 18th anniversary of our glorious elopement to beautiful downtown Reno. Every year we try to take a few days to recuperate from our jobs at a hotel within a few hours drive. As always happens, on the 4th day of our trip, I was barely well enough to make the drive home, and I spent the next week in bed. I'm still coughing now and and tomorrow I'll be mentioning it to my doctor.

I have a theory why it happens. Like many people, I am a stress junkie. Caffeine and adrenaline help me power through the work day. I just get more things done when I think the world will end if I don't get everything done. The stress keeps my immune system humming along. If I stop being stressed, I get sick.

There's only one problem with this theory. It's total BS. Wrong. Bogus. And I know it. But part of my brain, somewhere away from my frontal lobe, keeps telling me its true. But my frontal lobe is telling me that I probably have a confirmation bias.

Biases are very important to the sciences. That's primarily that you have to understand them in order to avoid or correct for them. The most famous bias is probably the "selection bias." Suppose you wanted to do a study on a bunch of random stars to see how bright the average star is. If you just went out at night and selected stars you could see that would be a poor choice, because most of those stars are extremely bright, as in thousands to millions of times as bright as the sun. If you did a proper selection of random nearby stars you would see that most stars are much smaller and fainter than the sun. Of course, doing it right is harder. The myth that the sun is an average, or even small, star probably resulted from this sort of biased study. The sun is an average star the way an A student is an average student. Both are brighter than 95 percent of the rest.

Biases of this type are often unavoidable. I've argued that the anthropic principle is a bias resulting from lack of knowledge about lifebearing planets apart from our own, including a lack of knowledge about whether they exist.

Other biases are built into our brains, and we couldn't live without many of them. There are lots of things that you don't want to waste actual thought on. For example, if you're driving and you see a yellow light, your frontal lobe doesn't ponder what's next. You've been biased to believe that a red light will follow a yellow one, and you, me, and everyone else will be totally unprepared for a malfunctioning signal where the light goes green-yellow-green-red repeatedly. When we're young, we're biased to be afraid of non-existent predators in closets or under beds.

Confirmation biases generally result because for reasonably rare events, we remember the events that occurred rather than all the times such events didn't occur. In my case, that would be the conjunction between vacation and getting sick. Do I remember all the vacations I took without getting sick? Of course not. Do I remember all the times I got sick without being on vacation? No. There's some possibility that my theory is right, but in order to decide if it's true I would need to keep detailed records.

Why won't my brain give up on this stupid idea? It's a survival skill. Replace "going on vacation" with "walking under a big tree" and "getting sick" with "being attacked by a leopard." Or similarly "eating smelly antelope meat" and "throwing up for 12 hours." People whose brains don't form that sort of attachment are probably at a disadvantage in an environment with large predators and potentially deadly food.

But it's not all good. Confirmation bias also has societal implications. For example, there's an incredible number of people who think violent crime rates are worse than ever when they've been falling ever since they peaked in the early '90s. A relative of mine said it doesn't matter if they're falling because it doesn't feel like they are. Ignoring contradictory evidence is a form of confirmation bias. Nobody notices a day when there isn't a violent crime in the neighborhood, but everyone notices when there is.

Of course, human beings aren't dumb. There are plenty of them ready to take advantage of your biases, even if you don't know you have them. Politicians, salesmen, advertisers, magicians, talk show hosts, poets, morning show DJs, comedians, that barista that hits on you every morning, medical quacks, home alarm companies, pickpockets, hollywood starlets and lots of other people make at least part of their living by exploiting your biases, some harmlessly, others not so much. Even scientists will use your biases to attempt to get you to click on a random link for no apparent reason.

So what does a scientist do about bias? Well, we need to know we have them and avoid or correct for them, because there's another scientist lurking around the corner waiting to catch your mistake. Publishing a paper explaining why someone else is wrong is one of the best parts of being a scientist and can be very rewarding in terms of publicity. But for the most part scientists are careful to practice "evidence based" science.

Sticking to "evidence based" practice in the rest of my life has worked pretty well. If someone tells me something I want to hear, the first thing I ask myself if why they would lie to me or if there is evidence for what they say. Cynical? Damn straight. I also prefer evidence based medicine, evidence based finances, evidence based politics and evidence based relationships. (That last one took me far too long to learn. Ask any old girlfriend.) Certain people should also stick to evidence based cooking.