Trying to Land a Plane (to Prove the Dunning-Kruger Effect)

Here’s a sanitized version of the provided YouTube transcript:

—

[Radio] I want you to go to 180- GFA 737, going to 180 speed. I am not a pilot. So what am I doing trying to land a 737 passenger jet? Okay, this is unusual.

(alarm sounding) – [Radio] (indistinct). – Oh, okay. (chuckles) Well, I’m trying to prove a point. Let’s keep the hands on the controls there, buddy. How big is the difference between what we think we know and what we actually know? Uh-oh, we’re going to miss the runway! (laughs) Well, we’re about to find out. Oh no, we’re in the air again. This is not good. (tense violin music)

Hey, smart people, Joe here. Let me know if this thought has ever crossed your mind when you’re sitting there on an airplane. Okay, what if the flight attendant suddenly got on the intercom and said, “Does anybody know how to land this plane?” What would you do? Well, I think about this all the time. And I think that I would actually raise my hand and say, “Yes.” Because there’s a certain kind of YouTube video that I love—it’s pilots landing planes, all those shots from inside the cockpit. I love this stuff. I cannot get enough of it. And you know what? It makes me feel like I could actually do this; I’ve watched enough of it.

So today we’re going to put that to the test by going to an actual flight simulator and see if I really know what I’m doing. Hey, so Future Joe here, that was me a couple of weeks ago. Probably a little more confident than I should have been. I know I’m not the only one who’s ever sat in their seat on a flight and wondered this exact thing, right? Especially if, like me, you played a lot of flight simulator games when you were younger and you love YouTube videos of pilots flying planes.

Seriously, they’re just so calming to watch; it’s like how I meditate. But I am not a pilot. And I think most of you watching this video, especially if you are a pilot, are probably wondering how could this guy possibly think he can actually land a passenger jet? Well, I did this because I wanted to understand how we judge what we think we know versus what we actually know.

Recently, some psychologists investigated this very question. When people watched a video of a pilot landing a plane, they were more likely to say they themselves could land a plane without dying, even though none of them had any actual flying experience, which is just nuts, obviously. I mean, flying a plane takes a ton of skill and practice. This is just one example from a huge body of research about a famous psychological phenomenon called the “Dunning-Kruger Effect.”

The “Dunning-Kruger Effect” is actually an entire family of effects, but the one that everybody knows is that people who aren’t knowledgeable or experts don’t know the depths of their lack of knowledge. If you’re making wrong decisions, you don’t know you’re making wrong decisions. My name is David Dunning, I’m a Professor of Psychology at the University of Michigan.

In 1999, he and Justin Kruger published a paper. They found that people who are incompetent at something overestimate their abilities more than anyone else, and the “Dunning-Kruger Effect” was born. It’s gotta be pretty strange to have something named after you that’s all about how people make mistakes or how ignorant they are.

Now, based on the countless hours that I’ve spent landing planes in Microsoft Flight Simulator or watching cockpit YouTube videos, I certainly wouldn’t claim to know as much as an actual pilot. I mean, flying a huge jet is an incredibly specialized skill, but I feel pretty confident that I know a lot more than the average person. But is that enough for me to bring a plane safely down to the ground with an expert pilot guiding me over the radio? Well, that is what we’re about to put to the test.

Hans: Hey Joe, nice to meet you.

Joe: I guess I’m going to try to land a plane today. Fingers crossed, that’s all I gotta say. (laughs) Hans is a pilot with thousands of hours of flying experience for a major airline. Now, if this were to happen in the real world, air traffic control would immediately bring in a pilot like Hans, who knows that particular aircraft, to direct a passenger on what to do.

I’ll be sitting in this simulator, a 100% accurate replica of a Boeing 737NG airliner cockpit. Hans will only be able to communicate with me via radio, and he’ll only be able to see what air traffic control would see—my altitude, where I’m heading, and how fast. He won’t be able to see any of the flight controls or what the plane’s doing, so this is really up to me.

So today we’re going to simulate a flight from Chicago to Boston in a 737 800NG. It’s a very common airplane; it has fully automated capability landing on the runway. The younger generation is very familiar with video games and so forth. You should be able to punch the buttons and program the airplane so it completely brings you down all the way down to the runway.

And then, maybe we’ll give me a shot at trying to do it.

Hans: Yeah, absolutely, though, because with the automation, I can follow you on the radar screens to see where the airplane is going, and then we’ll just have to talk you through making you land the airplane by yourself.

Joe: Wow, okay, there’s so much in front of me here. (chuckles) Hello, I’m a passenger on a flight to Boston without pilots. I need help.

Hans: That’s not a good situation.

Joe: So there have been cases where a passenger successfully landed a small plane after a pilot became incapacitated, but a passenger would likely never have to actually fly a modern airline jet. Not only do modern airliners have more than one pilot in the cockpit for this very reason, today’s aircraft are highly automated and many can fly and even land themselves if programmed correctly. If this scenario happened in the real world, this is what someone from air traffic control would direct a passenger to do.

Now comes the hard part.

Hans: All right Joe, we’re going to program the keypad on your right knee.

Joe: The screen near my knee controls the flight management computer. Hans is guiding me through a series of steps to give the plane the right information to land itself. By entering a series of GPS points along our route and information about the airport and runway where we want to land, this computer can control the autopilot all the way down to the ground. If I do all the steps correctly and remember to put the flaps and gear down, the plane should basically land itself.

There’s so much happening. And I don’t even feel like I’m actually flying; I’m just reading… The flying part, I don’t even know how I’m going to fit that in. As we approach the runway, the list of things I need to do seems endless.

Hans: The speed rate and move it one inch towards you.

Joe: I’ve seen those when you fly; they come up on the wing. That’s cool. Even though things are generally on auto, it’s not easy. This is quite a turn. Whew! Approaching the intercept. That sounded right; I don’t know. I’ve heard people say stuff like that.

Hans: I’m going to let you put the gear on.

Joe: That’s good. Gear’s good.

Hans: You should be able to see Boston.

Joe: Do you see the runway yet? Even though I can barely see the runway, the plane knows where to go. I can’t believe that this can just land itself; this is crazy.

[Cockpit Speaker] Approaching minimums. 100.

Hans: We’re going to (indistinct) here in like, 50.

[Cockpit Speaker] 50, 40, 30, 20, 10.

(Joe chuckles) (suspenseful music) (Joe exhales)

Joe: That was so much; I can’t imagine actually flying. (Joe exhales) My brain is exhausted from having to do so many things. And I don’t feel like I really actually flew the plane; it was just punching buttons and turning things. And I’m already overloaded; actually having to fly the plane, I don’t know how I’m going to have time to do it. There’s so many other things you have to do, like flying the plane? That’s too much.

So we’re going to try this again, and this time I’m actually going to have to fly. This is not going to go well. Okay, so I successfully got the plane on the ground. I feel pretty good; what could go wrong? I think I have manual control of the airplane.

Hans: Slowly push the yoke a little bit forward, so the flight director goes slightly below the horizon.

Joe: This is so hard. I’m having trouble keeping the plane at an altitude. I’m getting very low, actually; I need to pull up to 3000. I can barely make out the runway at this point.

Hans: Okay, that will be the runway you will be landing on, Joe, so we’re all counting on you.

(Joe chuckles) Okay. I can’t keep this thing on a course at all. A little bit more, a little bit more. I don’t feel like I’m lined up for the runway here.

Hans: You want to have the red over the white lights.

Joe: Yeah, definitely coming in low.

Hans: We’re going to turn off the auto throttle.

Joe: Okay, I have control of the throttle.

[Cockpit Speaker] 300.

Joe: Okay. Oh boy. Pull up.

[Cockpit Speaker] Approaching minimums. 100.

Joe: Uh-oh, (laughing) this is the runway! I’m going to hit these lights. Okay, we’re coming in over the end of the runway. Trying to power back.

[Cockpit Speaker] 20, 10.

(Joe exclaims) I’m like shaking. (chuckling) I don’t think that really would have worked.

Hans: Oh well, but you crushed the gear.

Joe: I crushed the gear?

Hans: Oh yeah. (both laughing)

Joe: In real life, this means the plane would have slid down the runway, likely starting a major fire. So technically, I crashed. Luckily, I’m going to get one more chance, only without the autopilot helping me at all.

So it turns out flying is harder than I thought, than just watching a few YouTube videos. When you actually have to fly the plane and put it on the ground, I didn’t do it, is what I’m saying; I crashed. Having a little bit of knowledge and seeing what pilots do in the cockpit from watching all these videos, it was this weird feeling of knowing I should be doing something but not knowing what I didn’t know how to do. And then all of a sudden there’s the runway, and it’s too late, and you break the landing gear off, and everyone’s in trouble. It didn’t go well, that’s what I’m saying.

So in some cases, having a little bit of experience may actually lead to more Dunning-Kruger overconfidence than having no experience at all. According to data, pilots with a couple hundred hours of flying experience tend to have the most fatal accidents, not pilots who are just starting out. It’s known in aviation as “The Killing Zone.” I feel like I’ve been there.

If you asked me, “Okay, who’s the most vulnerable to the Dunning-Kruger Effect?” I would say that everybody is vulnerable to the Dunning-Kruger Effect. And sooner or later, you will have your time in the Dunning-Kruger Effect. Though it’s likely the case, you won’t know. I think hitting the limit of your own knowledge is a day-to-day experience.

What Dunning and lots of other scientists have found is we are not always very good at judging our knowledge. And we overestimate what we know a lot. There’s countless studies, and the Dunning-Kruger Effect shows up just about everywhere we look. In one famous example, when given a quiz, more than 90% of people claimed to be familiar with completely made-up concepts when presented alongside real ones.

There are many reasons people may claim to know more than they really do. I mean, we want to fit in socially or we unconsciously tend to view ourselves as above average in everything from intelligence to sense of humor. In one study, 93% of Americans said they were safer and more skillful behind the wheel than the median driver, which is mathematically impossible.

But the Dunning-Kruger Effect is a special kind of overconfidence because it’s rooted in a paradox. You aren’t able to know what you don’t know, so when you mess up, you might not be able to realize it. The knowledge you need to come to the right decision is exactly the same knowledge that you need to decide whether the right decision has been made. And if you lack the first, you lack the knowledge you need for the second assessment.

In fact, the French novelist Anatole France once said a true education is not learning things but learning to separate what you know from what you don’t know.

Now, we’re going to up the ante a little bit, a nice cloudy day coming down into the runway. It’s down there somewhere. This is not going to go well. Auto throttle and autopilot are disengaged. Oh no, oh boy. If I breathe wrong, I’m off course here. Okay, I’m getting low. (Joe singing to himself) Just trying to stay right in here.

Hans: Let’s keep your hands on the controls there, buddy.

[Cockpit Speaker] Glide slope.

Joe: I hear ya. Glide slope.

[Cockpit Speaker] Glide slope.

Joe: Ah! Ah!

[Cockpit Speaker] 30, 20.

(Joe yelling) Oh no, we’re in the air again. This is not good. Wait, am I alive? The simulator doesn’t show it, but that was a crash, too. My score when actually flying is 0 for 2. That was not great. I apologize to the 150 or so people who paid for tickets on this flight.

So that didn’t go very well; the last time I crashed. That was just so much harder than I figured it would be, even understanding how a plane should be flown. And then I could tell there were just things that I didn’t know how to do.

Hans: So that’s what we call in aviation, Joe, the chain of events. We’re increasing the severity level slowly, making sure that you manually do the throttle and the yoke, and diverting attention to all the controls puts you in what we call an overload situation.

What we find with the Dunning-Kruger Effect is that people are unable to know what they don’t know, and that’s one of the differences between what experts are versus what normal people.

What does it mean to be an expert pilot?

Hans: To be an expert pilot is that we have had the training, so we know what to anticipate, what’s going to happen.

Joe: I knew that there was a problem, but I didn’t know what to do to correct the problem.

Hans: As a novice, you just don’t know what to expect. So every situation, every second is something that you have no clue what the outcome is.

And of course, being able to fly is an important prerequisite to getting to be a pilot, also, which I obviously don’t know how to do.

Hans: But just a little bit. You have potential; there’s room for improvement.

Joe: All right, so if you’re saying if this whole YouTube thing doesn’t work out…

Hans: There’s a place for you at the airlines.

(both chuckling)

Joe: What I’ve learned and what’s really the key here is the Dunning-Kruger Effect is not just that some people think they’re smarter than they really are. What it really says is that every one of us is ignorant about many things—economic theory, car repair, football strategy. That lack of knowledge means we won’t even know when we’re making mistakes.

You don’t know the things you don’t know, so when you mess up, you won’t even realize it. And it’s not just something that happens to other people; we’re all at risk of this in our daily lives.

You’re probably thinking what I’m thinking by this point, “But the internet.” Has all of this made the Dunning-Kruger Effect worse today? Well, this failure of self-knowledge has probably always been with us; it’s just part of the human condition. But the internet has made it more visible today for two reasons: we just interact with more stuff, so you’re going to run into more things you don’t know very much about. And the internet is designed to encourage you to talk about it.

The second reason the internet makes it easier to learn a tiny bit about something and enter that very dangerous Dunning-Kruger overconfidence zone, where you know just a little bit more than nothing.

So how do you avoid it?

One thing to do is to depend on other people, that they’re going to be perfectly knowledgeable as well, but two heads are absolutely better than one. Three are slightly better than two, so when a decision is an important one, you do want to check your math; you do want to check your reasoning with another person. If we try to go it alone, that’s when we get into trouble.

This is why pilots rely on those huge checklists and manuals. So they don’t rely just on their own knowledge in order to not make mistakes. And this is why science works; science outsources that whole process of checking for errors to lots of other brains. That’s what peer review means. Science doesn’t go out and confirm ideas; it tries to disprove them, and whatever survives is a little bit closer to the truth.

Confidence has benefits; it can be the boost that we need to help us take on life’s challenges. But overconfidence can definitely also be a bad thing. What we know is finite; what we don’t know is infinite. And that border between what we know and what we don’t know, we meet it more often than we think, even in everyday life. And that’s actually something we should all get more comfortable with because if we knew everything, life would be boring.

Not knowing things gives us a chance to learn, to be a different person, a better person tomorrow than we are today. Knowing how much we don’t know gives life a purpose. Stay curious, and please secure your tray tables to their upright and locked position; we’re approaching the end of this video.

Wait, land a plane with no experience? Surely you can’t be serious!

I am serious, and don’t call me Shirley. And surely we can’t end this video without thanking everyone who supports the show on Patreon; you guys help make videos like this one possible. If you’d like to see