Tag Archives: locomotion

The Broken Escalator Effect (It’s Real)

Every day we take the Paris Metro and RER to and from class. It’s a relatively painless trip, except when there’s a strike going on (which has been almost every day). One day last week, as our motley crew filed through our favorite station, Châtelet, to transfer trains, we reached our favorite stretch of the station: the moving walkways. I approached the walkway without hesitation, took a step onto the belt, and immediately felt myself jolted awake by a sense of falling. As it turned out, the moving walkway was broken that day, and pedestrians were just using it as a normal path. I followed suit and laughed silently at how funny I must have looked to anybody who saw me nearly fall on my face.

Our favorite Metro stop

Later that afternoon, on the return journey, I’d had ample time to wake up during the day. As we approached the same collection of moving walkways, I made sure to take note of the functionality of the machines. They were all still broken, but I decided to follow the crowd and walk along one of the belts anyway. This time, I approached, took a step, and felt jolted again! I was shocked at my brain’s miscalculation despite my conscious awareness that the walkway was stationary. I presumed that it had to be some sort of perceptual memory that I had for moving walkways. Perhaps because reality wasn’t matching up with what my brain had learned to be true of “people-movers” countless times before, my mind was having trouble adjusting. I decided it was worth a search in the literature when I got home.

No "broken escalator effect" here

What I found was not only reassuring for my vestibular system, but also immensely interesting. There is an extensive collection of scientific research on what has been called the “broken escalator phenomenon,” (Reynolds and Bronstein, 2003). Evidently, the effect is more evident on moving walkways, but because nobody knows what to call them, the original authors of the phrase decided to go with escalator instead. Once the phenomenon became well known as a common occurrence in city-dwellers, researchers sought to describe what was actually happening to cause this “feeling of uneasiness” despite absolute consciousness of the fact that the conveyor was not moving

First, experimenters had subjects walk on a short, stationary moving walkway a few times while measuring walking speed, postural sway, and muscle contraction (Reynolds and Bronstein, 2003). Afterwards, the experimenters turned on the walkway and had the same subjects board the machine. Not surprisingly, subjects made several physical changes as they got used to the moving version, but the most significant change observed was in the actual velocity of movement just prior to boarding. Naturally, the subjects increased their pace by .3 m/s in order to minimize being jerked by the belt. This is similar to what happens to us in everyday life. We encounter a majority of moving walkways in their “on” position, and we become accustomed to increasing pace, leaning forward, and flexing our leg muscles as we approach them. Next, the researchers informed the subjects that the walkway would be turned off, and in fact, they could see so for themselves. When they approached the walkway this time, all subjects stumbled, and many were shocked or laughed at the occurrence. Analysis of the physiological data showed that approach velocity, trunk lean, and muscle contraction took place at levels in between normal walking values and the values seen when subjects were accustomed to the moving walkway. It seemed that the brain was confused by seeing a normally moving pathway in a motionless state, and addressed the situation by “hedging its bets” so to speak. Interestingly, repeating a second trial with the “off” walkway shows no signs of distress. The brain learns quickly to adopt normal walking motor programs for the motionless walkway. Further studies have shown that skin conductance also increases just prior to experiencing the “broken escalator phenomenon,” implying that subconscious, fear-based mechanisms are at play (Green et al., 2010). This may explain why the hiccup occurs even when one consciously recognizes that normal walking will suffice.

Primary motor cortex, where the researchers stimulated.

Given that this phenomenon is strikingly similar to the lack of balance that many neurological disease patients experience, further studies aimed to find ways to modulate to the occurrence (Kaski et al., 2012). Recently, researchers tried this using a technique called transcranial direct current stimulation (tDCS), which is a lot like connecting a battery to your skull, except it’s scientific. Subjects went through the same experimental procedure as in the first study, but just before had a small anodal current passed through their brain for 15 minutes before the moving platform phase of testing. The researchers targeted the primary motor cortex, an area of the brain responsible for executing movement and storing motor memories, or the actual plans that the body uses to coordinate movement. The researchers believed that the broken escalator effect occurred due to an inability to suppress the brain’s default “moving walkway motor plan,” so activating primary motor cortex would cause the phenomenon to become even more extreme. Indeed, the subjects who received the electrical stimulation showed a larger broken escalator effect and took more trials to adjust to the stationary pathway than control subjects who received no stimulation. Though the nature of the experiment did not necessarily prove that the broken escalator effect is due to overactive motor memory, the results are significant in that they show it is possible to manipulate gait and motor problems with relatively simple technology. tDCS is fairly cheap and straightforward compared to other similar technologies, and its lack of precision actually lends itself nicely to working with the distributed neural systems of locomotion. Though this study used tDCS to worsen a locomotor problem, this same system may soon become a useful tool in neurological diseases that show locomotor symptoms such as stroke, Parkinson’s multiple sclerosis, and Alzheimer’s disease.


-Max Farina



Reynolds RF, Bronstein AM (2003) The broken escalator phenomenon. Experimental Brain Research.

Green DA, Bunday KL, Bowen J, Carter T, Bronstein AM (2010) What does autonomic arousal tell us about locomotor learning? Neuroscience 170: 42-53.

Kaski D, Quadir S, Patel M, Yousif N, Bronstein AM (2012) Enhanced locomotor adaptation aftereffect in the “broken escalator” phenomenon using anodal tDCS. Journal of Neurophysiology 107: 2493-2505.