Let’s face it: Our society rewards persistence. We tell squirming children to just be patient. We resent people who get to skip lines. We praise those who “stick with it”—no matter what “it” is.

But is waiting always worth it?

Let’s say that you’re in a bit of a rush, and you just called a car using a rideshare app. The app shows you how far away the driver is and how long it will take for them to get to you (10 minutes). As time passes, the clock counts down, and the driver gets closer. It might feel annoying to wait, but it makes sense to wait. After all, if you weren’t willing to wait 10 minutes, why would you have accepted the ride?

But now imagine a slightly different scenario, which may also be familiar: You accept the ride, and the app tells you that the driver will be there in 10 minutes. But as you look at the map, you see the driver’s car’s avatar start to spin in a circle. Then it seems to be going in the wrong direction. Then it seems to be stopped. The estimated time still says 10 minutes, despite the fact that you know a few minutes have passed. Now, what do you do? Many people in this case would give up on waiting. They would cancel the ride and try another app or another mode of transportation. Is that irrational? Are those quitters lacking in self-discipline? I think most of us would say that they are being reasonable.

Our Expectations

What the above thought experiment demonstrates is that whether or not persisting for some reward is worth it depends on your expectations about how long you will have to wait. In the first scenario, you saw that the car was getting closer and closer, so you expected that the wait time remaining was decreasing. In the second scenario, however, there was some uncertainty about when the car would arrive.

In these cases, people tend to have “heavy-tailed” expectations. This essentially means that they expect the extremes: Sure, there’s a chance that the car will come right away, but there’s also a good chance that the car might not show up for a very, very long time. Another example of when people have these heavy-tailed expectations is when they wait on hold with customer service. In the absence of clear indications of when you will get to talk to someone, you will probably hang up after waiting for just a short amount of time, because if they don’t connect you right away, well, they might not connect you for a very, very long time.

It turns out that, when people have heavy-tailed expectations about how long they will have to wait, it is actually smart for them to quit waiting after just a short while. Why is that? This is explained in detail in a 2013 paper from Joseph McGuire and Joseph Kable, but, essentially, there is an opportunity cost to waiting. As you wait for the car that might not come for a long time, you are missing out on a chance to use other transportation. As you wait on hold, you are missing out on the chance to do other things. So, even though “quitting” gets a bad rap, the savviest among us know what situations call for giving up.

New Study

In a study I conducted with Kable and other colleagues, participants did a computer task in which they collected virtual “tokens” that were later converted to real money. They had 10 minutes to gather as many tokens as they could, but they could only gather tokens one at a time, and each token took a while to appear. Participants had no idea how long each one would take to show up. For some tokens, participants might only have to wait two seconds, but for some, they might have to wait 40 seconds. Our subjects did have the option to quit waiting for any given token at any time, though. Then they could move on to a new trial,​ which might have a shorter wait time.​

Our goal in the study was to see whether, under conditions where there is uncertainty about when outcomes will materialize, people’s expectations are “heavy-tailed”?

To get an objective marker of expectations, we measured participants’ pupil dilation. Since people’s pupils dilate when they are surprised, we could measure how surprised people were to get the token when it arrived. It turns out that the longer that someone had been waiting for the reward, the more surprised they were when it finally showed up. This might seem counterintuitive; after all, you know the token will arrive at some point, so the longer you wait, shouldn’t you be less and less surprised when it does show up? But this result—more surprise after waiting a longer time—is completely consistent with people’s expectations being heavy-tailed. Basically, people were surprised when those tokens eventually showed up, because they had already started thinking, “Well, it’s probably going to take forever.”

The key takeaway from our study is that, when people don’t know how long they will have to wait for something, they tend to expect the extremes: Either this thing will happen right away, or it will take forever, and it’s not worth waiting. And when we have those expectations, it is actually rational to quit waiting after a little while, so that you don’t waste too much time.

In other words, patience is not always a virtue, because some things are simply not worth waiting for.