Basic Biases: The Availability Heuristic

The availability heuristic is a bias that arises when we confuse probability with ease of recall. This means that without noticing it, we are actually answering a completely different question than the original one. Instead of answering "how likely is this?" we answer "how easily did this come to mind?". If our experiences about the world would be uniformly and randomly distributed – and covering all possibilities – only then would ease of recall be the same as probability. Of course, this is not the case. With modern media, private experience is not the only source for our thoughts; we read newspapers, blogs, watch TV and consume all kinds of media that tell us what did, could have, or should have happened.

My favourite example of the availability heuristic is related to travelling. Imagine that you’re finally getting to that long-awaited holiday on a paradise island. Your friend drops you off at the airport. As you gather your suitcase and are just about to leave, your friend shouts to you “Have a safe flight!” You say thanks, and proceed to check-in.

Why this is a good illustration of the availability heuristic is the fact that you – the one getting on the plane – is being reminded to have a safe trip. Whereas in fact, if you look at the numbers, driving a car is actually much more likely to be fatal than flying! The discrepancy is huge: statistical estimates say that it’s around two to five times more likely to have an accident on the way home from the airport than on the flight, though the exact numbers depend a lot on the assumptions of who's driving where. So when we consider the safety of the car versus the plane, it’s very easy to remember examples of planes crashing, or even disappearing altogether. Whereas car accidents are so common that they rarely break the national news barrier.

So it’s not just that availability is a poor guide to probability. In the case of mass media, availability is actually inversely proportional to the probability! After all, papers want to report new, exciting things, and not just car accidents that happen every day. This essentially means that “oh, I saw an article about this in the paper” is not a good guide to the world of things to come.

If you’re a nitpicker (I know I am, so there’s no shame to admitting) you might say that saying safe trip is really not a probability estimation claim. When you say “have a safe trip”, you’re not trying to state that “I believe your mode of travels is statistically more likely to result in death or injury, and I aim to prevent a part of that by this utterance”. No, of course not. Even an economist wouldn’t claim such a thing! It’s more a statement of wishing your friend well, and hoping for a good trip. But still, I find it funny that we use the word “safe” here, in exactly the wrong place.