Roman Mars (RM): This is 99% Invisible. I’m Roman Mars.
RM: If I asked you to close your eyes and mimic using one of the simple human interfaces in everyday life, you could probably do it. No problem. Without having a button to push, you could close your eyes and pretend to push a button, and that would accurately reflect the action of pushing a real button. Same with flipping a switch or turning a doorknob. If you close your eyes and fake the movement, it would sync up with its real-world use.
RM: Now if I asked you to do the same thing with a car’s steering wheel, you think you’d be able to describe steering accurately and mime the correct movements with your hands in the air, but you’d be wrong. So wrong. You’d probably kill a bunch of imaginary people. But don’t just take my word for it.
Dr. Steve Cloete (SC): My name is Steve, I’m a human factors engineer.
RM: His full name is Dr. Steve Cloete, he just introduced himself as Steve. He’s a casual guy.
Tristan Cooke (TC): And what was your research on? What was the main research you’ve done?
RM: And that’s Tristan Cooke and Thomas Nelson is also in the room there asking questions of Steve. Tristian and Tom run this awesome blog called Humans in Design.
SC: My interests are in driver behavior.
RM: So here’s the exercise. Put your hands up and grip an imaginary steering wheel. Ten and two.
SC: So I’ve got my hands at ten and two, yeah.
RM: Now I want you to mime the action of changing lanes to the right. So have you tried it? Alright, if you’re like most people, according to Steve’s research, we’re talking the vast majority of people, you turned your imaginary wheel 45 degrees to the right or so, and then back to the center. And that’s the action for changing lanes. But. That’s not right.
SC: What you’ve described is actually a sequence of movements that you require for turning a corner.
RM: To actually change lanes, you have to turn the steering wheel to the right, then turn it back, crossing the center line, turning the wheel in equal amplitude to the left to correct your heading, and then back to the center. Knowing that this weird brand hiccup existed, Steve’s team at the University of Queensland tested people in driving simulators. And sure enough, if you provided visual feedback, people could change lanes no problem. But as soon as you turned off the screen or blindfolded them and told them to change lanes, even though they had only just moments ago successfully performed the correct action, they couldn’t do it.
SC: That produced pretty catastrophic errors.
RM: Given no other feedback, you can’t change lanes using a steering wheel without vision. I mean, since cars are designed to be driven with your eyes open, that’s okay. In fact, this is a funky example that doesn’t mean that a steering wheel is a bad interface design. You can probably argue the opposite, you could argue that a steering wheel is so well-designed that anyone can use it with just a little bit of feedback in the real world, even though they can’t describe it in the abstract.
TC: So people didn’t learn?
SC: Not over the course of an experiment, which takes about 45 minutes to an hour. The size of the error was pretty consistent.
RM: But Steve and his researchers decided to test this in a real car, so Steve got something like 50 million dollars worth of insurance and took people out in a real car and drove them down a test road and then took their vision away, and asked them to change lanes.
RM: Our friends Tristian and Tom were two of the participants.
TC: The first time you look vision away from me it was terrifying.
RM: It’s a little bit like when Obi-Wan Kenobi was teaching Luke Skywalker how to use a lightsaber without using his eyes.
“With the blast field down I can’t even see!”
RM: In the first few runs, people made the same mistake they did in the simulator- essentially they were told to change lanes and they moved the steering wheel in such a way that they turned a corner instead. But here’s where things get really interesting.
“Stretch out with your feelings!”
RM: People gradually started to improve. And by the end of the real world blind trial, they were performing almost perfect lane changes.
“See? You can do it. “
“I call it luck.”
RM: In my experience, there’s no such thing as luck.
SC: When you’re driving a car, you have a whole bunch of extra visual information you can use apart from vision.
RM: There’s the vestibular system that senses your balance, or the somatosensory system that senses pressure on your skin as your thigh presses into the seat as you turn a corner.
TC: One gives an erroneous signal, the others would correct it?
SC: That’s right.
RM: Using all those extra senses, people learned how to perform the lane change in a real car, blindfolded. But here’s the crazy part. After it was over, they asked the participants to mime a lane change with an imaginary steering wheel in their hand. And sure enough, they went right back to making the same mistake, performing a corner turn instead of lane change.
SC: Just how robust this error was. And how it was resistant to learning.
RM: The steering wheel is pretty nicely mapped to our intuition, but it still requires the slightest bit of feedback so that our brain knows how to use it. But we still manage. Also, I can’t stress this enough, people, don’t drive blindfolded. Really, you’re not going to be able to do it. Alright, I’m glad we had this chat.