Monthly Archives: April 2016

Understanding Bias – What color is this truck?


No, it’s not a trick question.  Go ahead and say it.  The truck is yellow.

So what’s the point?  How does this help us grasp the concept of cognitive bias?  Well, consider this quote from David McRaney,

“If you notice a rise in reports about shark attacks on the news, you start to believe sharks are out of control, when the only thing you know for sure is the news is delivering more stories about sharks than usual. . . you think, ‘Gosh, sharks are out of control.’ What you should think is ‘Gosh, the news loves to cover shark attacks.”

It’s actually very reasonable to assume that the more you hear about shark attacks, the worse the problem is. It’s how our minds work.  Our brains are making an assumption that works really well in small-scale.   In other words, when our world is small enough that the stories we hear are representative of the environment we experience, then the assumption will most often lead us to a useful conclusion.  If we hear that numerous people have been attacked by sharks down at the local swimming hole, we’d do well not to go swimming there.  However, if the context is big enough to include beaches along all the world’s oceans, then noticing an increase in reports may not indicate any particular risk to us, and assuming that it does leads us to a conclusion that is erroneous rather than useful.

The mental shortcut we use by making this assumption is an example of a heuristic. (specifically, Availability heuristic)  Since our brains are used to working with only partial data at best, they have become quite adept at using limited information to construct a bigger picture.  It’s very helpful much of the time, but not all of the time, because heuristics are limited in how accurate a picture they can provide.  There is a built-in chance for error.  That’s why the other component of many heuristics can become a problem; they’re automatic.  A shortcut becomes less useful if you have to take the time to stop and remind yourself to use it, and our brains value efficiency.  So, our brain often sees fit to make assumptions without conscious bidding, which allows them to remain hidden and leaves the door wide open for bias.

So what does all this have to do with the picture?  Well, first understand that saying the truck is yellow is not biased.  It’s an example of inductive reasoning, which is different. However, there are similarities that I think can help us grasp the concept of heuristics/bias.  Considering the limited information you have been given in the picture, all you can say with confidence is that this side of the truck is yellow.  Your brain, however, automatically used the data to construct a bigger picture, and in that picture the entire truck is yellow.  There is nothing wrong or even sloppy about making this assumption.  It’s almost certainly going to be correct, but there is a chance that it isn’t.   The thing to notice though is that not only did you make the assumption automatically, you remained unaware of it until it was pointed out.  This is similar to how a cognitive bias works.  Making the assumption is the heuristic, and the fact that you were blindly inclined towards an answer that could be wrong is the bias.

The next thing to notice is that, despite understanding the assumption and seeing the possibility of error, you will still continue to think of this truck as being entirely yellow.  You may be able to entertain the idea of it being black or blue on the other side, but you still would put your money on yellow.  Yes, that’s because you have good (inductive) reason to, but try not to miss the point. This is, again, similar to bias.  Just because you recognize that your answer could be wrong, that it could be biased, that doesn’t allow you to resist the inclination that it feels right.  Simply becoming aware of a bias is not enough to shield you from it. (see cognitive bias mitigation)  Essentially, your brain thinks it has good reason to reach the conclusion.

You may even be able to condition yourself to stop and consider the color of each partial car you see. You may be able to prevent yourself from making the assumption until you have more evidence. However, that doesn’t mean you have exiled such assumptions from your mind. The next time you are driving along a street, you will assume the houses are the same color on the sides you don’t see as they are on the side that you do.  And this is another similarity to biases.  You maybe able to train yourself to avoid a bias in very specific situations, but the very same bias can easily creep back in if the details are changed.  (see Wason selection task)

Now, please don’t go around questioning your assumptions about the colors of cars or houses.  That would be exhausting. And remember, this is not an example of some new “color” bias. The point is to provide a framework for understanding how heuristics work and how they can manifest as biases.  If you can grasp the mechanics of inductive reasoning, you can apply them to the concept of bias.

So please do keep this example in mind when you consider something like, for example,  confirmation bias.  Being aware of confirmation bias does not help you to avoid it, it’s part of your cognition, and if you want to make the best attempt at weeding it out of your conclusions, you’ll have to reach outside yourself, beyond your own cognition, and use a process designed to compensate, like the scientific method.

Much thanks to the blog Bias and Belief and the author’s response to my original draft of this post for helping to improve it.