Nudging to save lives!

A good friend Nirajan Rajadhyaksha sends this superb article which shows how behavioral insights are being used to save lives in Mumbai. (Nudges Blog pointed to an earlier article as well)

People in Mumbai usually cross railway tracks in order to get quickly on the other side, leading to many accidents and loss of lives:

The suburban rail system in the Indian megalopolis of Mumbai is best visualized as two slim arteries cutting through a crowded peninsula. On a map, the Western Line runs due north; the Central Line begins similarly, then wanders away into the city’s northeastern parts. These two lines and a couple of adjunct capillaries, making up a rail network dating back to 1857, carry roughly 7 million commuters a day, some of them over distances as long as 75 miles.

Every mile of this network runs through dense pockets of population, houses, and buildings; these are often just yards away from the tracks, separated at best by a low wall. Sixty percent of the length of the Central Line, for instance, has slums on either side. At rush hour, trains barrel through every couple of minutes, and pedestrian bridges over the tracks are rare. As a consequence, the most popular way for pedestrians to get between east and west Mumbai is to dash illegally over unguarded sections of the tracks.

Even where there are pedestrian bridges people prefer to cross changing trains etc. For instance, in Mumbai you have both slow and fast trains. Former stops at all stations and latter at few selected stations. So people change tracks as say a slow or a fast is late/cancelled. Here the trains are hugely crowded and people travel a lot of distances. Hence, there is always pressure to be quick to reach workplace on time. Uusal rational approaches like Caution signs and warnings do not seem to be working as well. In many stations, there are inspectors who charge a stiff penalty and even put you into a jail for crossing, but still people cross. Imagine the pressure.

Hence, there are experiments to prevent people from crossing these tracks.

For just over a year, however, an experiment at Wadala station, on the Central Line, has been hinting at unorthodox solutions to this problem. On the surface, the experiment involves small, odd changes. Certain railway ties have been painted bright yellow; a new kind of signboard has been installed near the tracks; engine drivers have modified the way they hoot their warning whistles.

This modest tinkering has had dramatic results. In the six months before the experiment went live in December 2009, Wadala had recorded 23 track-crossing deaths, said M. C. Chauhan, a manager with the Central Railway’s Mumbai division. Between January and June 2010, that number had dropped to nine; in the next eight months, up until February 2011, only one death was registered. “We think the project is a huge success,” Chauhan said.

The experiment is a pro bono safety project conceived by a Mumbai-based “behavior architecture” consultancy named Final Mile, which uses the lessons of cognitive psychology to influence people on the brink of making decisions. Classical economics has long held that human beings are largely rational, even as a century’s worth of psychological study has suggested otherwise. Advertisers and marketers have crafted campaigns on the premise of the rational being, believing that, say, a detergent brand need only insist on how much cleaner it can clean. But recent studies in behavioral economics and cognitive neuroscience have emphasized that human decisions are fraught with irrationality — that a detergent buyer may be more influenced by the shape of the container, or where it is stocked in the store.


Three interventions by the agency. Pictures are here.

First, Final Mile painted alternate sets of railway ties in fluorescent yellow — five painted, five unpainted, and so on — to tackle what is known as the Leibowitz Hypothesis. As laid out in a 1985 issue of American Scientist by experimental psychologist Herschel W. Leibowitz, the hypothesis found that we frequently underestimate the speed at which large objects move. Leibowitz, who died earlier this year, first observed this with aircraft, and in 2003, a pair of scientists proved the hypothesis for trains. “The misperception happens because the brain has no frame of reference, no way to evaluate roughly how fast a train is moving,” said Satish Krishnamurthy, a Final Mile behavior architect. But with the new paint job, Krishnamurthy said, “the mind now has a way to gauge the train’s speed, by observing how fast it traverses these ties.”

Second, the consultants replaced the stick-figure signboards with a graphic three-part tableau, featuring in extreme close-up the horror-struck face of a man being plowed down by a locomotive. “We hired an actor,” Krishnamurthy said, smiling, “because it had to be realistic.” They were drawing on the research of Joseph LeDoux, a New York University professor of neuroscience and psychology. LeDoux studies the links between emotion and memory, and in particular the mechanism of fear. “Emotional memory is stored in the nonconscious part of your brain,” Dominic said. “If you’ve been in a car crash and, months later, you hear tires squealing, your heart rate goes up and you start to sweat. That’s because your emotional memory has been stirred up.” The new signs dispense with explanatory text and instead attempt to trigger an emotional memory of fear.

Final Mile’s third intervention required train drivers to switch from one long warning whistle to two short, sharp blasts. By way of explanation, Dominic cited a 2007 paper from the Stanford University School of Medicine, which found that brain activity — and hence alertness — peaks during short silences between two musical notes. “The silence sets up a kind of expectation in the brain,” said Vinod Menon, the paper’s senior author and a behavioral scientist working with the Stanford Cognitive and Systems Neuroscience Lab. “That’s the way it works in music, and it isn’t inconceivable that it would work similarly with train whistles.”

It has been a major success an is being extended to whole train line in Mumbai (50 cr is being spent on this) and has lessons for others as well:

These simple, inexpensive interventions have worked so well that they’re now being extended across the length of the Central Line. But the larger implications of the experiment stretch beyond Mumbai, and beyond track-crossing deaths as well.

India is not the only country with a track safety problem. The very nature of railroads — tracks stretching many hundreds of miles, often into rural territory — has meant that the trespassing problem is difficult to solve. In the United States, the number of pedestrians who were fatally hit by trains rose 8 percent from 2009 to 2010, according to the Federal Railroad Administration Office of Safety Analysis. California, Florida, and Illinois rank at the top of the fatality data set; the Massachusetts Bay Transportation Authority, by comparison, fares well. In California in 2010, 66 people died after being hit by trains; the MBTA’s corresponding figure is only 2.

What more does one want from behavioral sciences. It is now being used to save lives as well…

As this blog has been saying, India should urgently have an office (minus the usual bureaucracy) looking at behavioral insights to shape/change things. There is so much which can be done.

3 Responses to “Nudging to save lives!”

  1. Tarun Says:

    Wonderful….thanks for sharing…

  2. Nudging to prevent drunken driving and accidents « Mostly Economics Says:

    […] like the way Central Railways used the nudge ideas to prevent people from crossing rail tracks which showed some great […]

  3. ashwariya Says:


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: