I need to set this article up with a story about an escalator I deal with on a daily basis. So if you want to skip the rant and get to the science, click here

I work in an office complex with one of the most backwards escalators in existence. I do not make this claim lightly—I have thought long and hard about why this escalator is so terrible. I’m going to preface this description with a big “First World Problem” warning, because I know one of you is already thinking it.

First of all, it’s the only way to get to the office cafeteria (I see a thousand mouse cursors already inching towards that “skip” link above) If the escalator breaks, they have a security guard ferry people down on the elevator. But if there was ever a fire during escalator repairs, well, sorry folks. Just kidding, they only ever pull apart one lane at a time, probably for this reason. Safety first, right? We’ll get to that issue a little later.

Second, the designers, in their infinite wisdom, made the most highly trafficked  part of the building (for all the tenants, the cafeteria is the one destination they have in common) only wide enough for one person, so if you get stuck behind somebody, you’re stuck. This particularly irritates me because I’m an escalator walker, and I don’t understand why more people aren’t.

Let me take a detour for a moment, and allow me to preach to you this one truth: The single greatest thing about escalators/moving walkways is the power of super speed. Because you have the power to move at the combined speed of the machine and your own pace. It’s fantastic! There’s no feeling quite like walking briskly on a moving walkway and watching everybody fall behind into your mechanically assisted dust. This sort of man/machine union can only be found in the operation of a basketball trampoline or in the super wee.

For many people, this is their only chance to be 1/10th of an Usain Bolt, and they squander it, content on passively letting the escalator/walkway take them to their destination. Lazy? Safe? To walk or not to walk? I may rail against the escalator standers, but I will be the first to admit the issue is more complicated than it seems. Economists have lost sleep doing the cost/benefit analysis of escalator standing, and in some places it is expressly prohibited to walk due to the risk of becoming drunk with Barry Allen-like abilities and falling on your head when it comes time to disembark. I sympathize with the concern—have you seen the edges of escalator steps? Vicious, serrated blades. In a perfect world, my office building would have an egalitarian escalator, where those drunk with power can grip life by the throat and vault into the gleaming sun like Icarus in a button-down shirt, and the rest can stand aside and move at the standard escalator pace of 0.30–0.61 m per second.

Thus far we have two strikes against the escalator of unnamed-office-complex on Riverside Drive in New York City: No alternative route and no space to move past slow people. We now come to strike three. Earlier, I called this escalator “backwards” and perhaps those of you who love a good Agatha Christie novel may have recognized that I was foreshadowing. You see, the escalator is reversed. The lanes go in the wrong direction. Need a visual cue? Observe fig. 1

Photo by JD Lasica

Perhaps if I worked in the United Kingdom, where one drives on the left side, I would understand the odd choice of lane placement. But it makes no sense in its current setting. It’s actually kind of a disaster. Not only do you need to pass across the “exit” lane in order to go down to the cafeteria, people are expecting to travel in the right lane (again, driving*) and they round the blind corner expecting to enter the correct lane. In both cases, they end up colliding with a plate full of spaghetti and meatballs.

The only explanation for this backwards design is that upon paying for your food, you snake your way through a narrow hallway and end up on the left side of the escalator. Ah, the light bulb clicks on. The designers clearly didn’t want people leaving the cafeteria to cross the people entering, and instead want you to get all the way to the top before spilling your lunch all over some poor old woman. On top of that, the hallway is so nondescript that people rarely even use it.

Disaster. Let’s get to the science, shall we?


In many ways, learning to walk is learning to fall. It’s a repeating cycle of controlled falls and self-catches. Once learned, it is performed automatically without thought, requiring very little conscious attention to obstacles, terrain, or  even shifting surfaces. When you think about it, it’s amazing we can go an entire day without tripping, stumbling, or falling all over ourselves, being the clumsy, top heavy creatures that we are. So imagine my surprise when I found myself running smack dab into a glitch in our automated transportation system, caused by another such automated system.

The other day they had shut off the terrible escalator in my office building. Normally they close the whole thing off, but this time they took a hint from the late Mitch Hedberg and allowed the broken escalator to become temporary stairs. Upon stepping onto the escalator, I found myself gripped with a strange sensation. It was a sudden vertigo that continued almost the entire way up. My eyes were saying, “It’s not moving.” but my sense of balance was screaming otherwise. The thought struck me immediately: I knew the escalator wasn’t moving, yet something in me was compensating for inertia that wasn’t there. Puzzled, I began to look around for this curious effect, finding a goldmine in what became known as the “Broken Escalator Phenomenon” (BEP from here on out).

Similar to the “Empty Suitcase Superstrength” and “I Could Have Sworn There Was One More Step” phenomena, BEP seems to be a product of expectation versus reality. But the former phenomena are caused by unknown variables—suitcase weight and number of steps on a staircase—while with BEP, one is 100% aware that the escalator isn’t moving. That’s crazy.

In 2003, Adolfo M. Bronstein, Karen L. Bunday, and Raymond Reynolds**, intrigued by anecdotal accounts of BEP, decided to put it to the test. 14 subjects were asked to step onto a moving walkway from a stationary platform. Once subjects became attenuated to the speed of the walkway, they were asked to do the same after being told that the walkway was turned off. What Bronstein and his team observed in the subjects’ gait was an increase in speed (to match that of the walkway) and what they called a “trunk overshoot” (the act of leaning forward to compensate for the movement of the walkway). The subjects were predictively stepping onto the walkway as if it were moving, even when fully aware that the walkway was not in operation. This stumble came to be classified as a locomotor after-effect (LAE). They  put it like this:

 …it demonstrates dissociation between the declarative and procedural systems in the [central nervous system]. Since gait velocity was raised before foot-sled contact, the findings are at least partly explained by open-loop, predictive behaviour. A cautious strategy of limb stiffness was not responsible for the aftereffect, as revealed by no increase in muscle cocontraction. The observed aftereffect is unlike others previously reported in the literature, which occur only after prolonged continuous exposure to a sensory mismatch, large numbers of learning trials or unpredictable catch trials. The relative ease with which the aftereffect was induced suggests that locomotor adaptation may be more impervious to cognitive control than other types of motor learning.

In other words, it seems that when it comes to walking, the brain is fairly quick to adapt to changes in terrain, and it does so with little to no conscious effort. After only a few trials, subjects were able to adjust to stepping onto a moving surface with ease. The downside to this “hands free” sort of learning is that when the terrain is changed again, no amount of conscious effort will keep you from stumbling. Despite being told the walkway wasn’t moving, having to confirm that they understood, and even being able to see that the machinery running the belt wasn’t moving, the subjects stepped onto that walkway as if it was moving.

Over the next few years, many follow-up trials would be performed to try and pick apart why our body reacts one way, despite full conscious knowledge to the contrary. These trials took me down a rabbit hole1 of experiments and exercises in “sidewalk science”  ranging on the use of split-belt treadmills to test gait adaptation, wooden stairs that mimicked the uneven steps of a stopped escalator to see if that was what caused LAE, to the use of labyrinthine-defective (vertigo) subjects to determine how much the vestibular system played a part in it all.

The result? Varied. But what seems to be clear is that when it comes to walking, we have a very sophisticated, anticipatory auto-pilot. Subjects had little trouble adjusting to changes on the split-belt trials, and were able to learn and unlearn gaits relatively quickly—even when distracted. In the 2008 Bunday/Bronstein follow-up, it was discovered that while a having a balance disorder made “braking” on the LAE (the amount of time it took to correct the trunk overshoot) it had little effect on the earlier learning process. Curiously, while vision had no effect in the same study, the 2009 Japanese study found that the wooden stairs made to look like a stopped escalator induced no such LAE. Whether this changes if the subjects approach the “escalator” blindfolded was not studied. I’d be curious to know whether the effect is caused by the subconscious “knowing” of what an escalator is, or if it is the sight of one.

It’s incredible the amount of research that can be gleaned from this one phenomenon, and what it is teaching us about how the brain works. All these studies show that we not only learn to fall in a controlled manner, but that we’re constantly updating, learning, and relearning how to do it as our situation requires. Inevitably, as creatures of habit, we find ourselves tripped up by small changes in the schedule mapped out by our feet, even if the reality is right in front of our eyes.


 This post was brought to you by Lanthanum (La).


*It is my anecdotal observation that since many New York City residents do not know how to drive, they similarly do not know how to walk. So they do things like walking slowly on the left side of a busy walkway/staircase and then make a right turn, cutting across everyone. When driving, the left lane is for passing, turning left, or incoming traffic.

**Sorry. Paywalled. You can read the authors’ follow-up in full here.

 1The more I looked, the more I found, I quickly found myself in over my head in reading. If you have any relevant information that should be added, please let it be known in the comments.