Behavioural solutions for road safety (Mint)

Posted on Published by Anand Damani

Behavioural solutions for road safety

This editorial article first appeared in Mint on 21st March, 2017

Making roads better should reduce the number of accidents. Yet that’s exactly the opposite of what’s happening in India. Despite measures being taken by the government on improving roads, there has been a continuous increase in road crash deaths since 2007, with a brief annual reduction in 2012. Between 2010 and 2015, incidence of road accidental deaths increased by an annual average rate of 1.2%. There were over 500,000 road accidents in 2015, up from 489,000 in 2014. More than 500,000 people were injured in road accidents in 2015, up from 493,000 in 2014. A total of 146,000 people died in road accidents in 2015, up from 139,000 in 2014. According to the National Crime Records Bureau, out of 146,000 deaths, only 0.8% of the cases were due to lack of road infrastructure.

Road safety is not just about creating infrastructure. It is about designing behavioural solutions that take human biases and irrational behaviour into consideration. When the roads are smooth, wide and empty, drivers are likely to speed. If the car being driven is big and tough, the driver feels much safer compared to driving say, a small hatchback. That makes drivers over-compensate and take undue risks. Regular speed limit signs are ineffective at getting drivers to slow down, because drivers don’t choose the speed based on speed limit signs. Rather, drivers simply go with the flow depending upon the width and smoothness of the road and traffic conditions.

To get drivers to reduce speeding, there have been several effective behavioural design nudges implemented around the world. At the curve of Chicago’s Lake Shore Drive and Oak Street, a series of horizontal white stripes have been painted on the road, that get progressively narrower as drivers approach the sharpest point of the curve, giving them the illusion of speeding up, and nudging them to tap their brakes.

According to an analysis conducted by the city’s traffic engineers, there were 36% fewer crashes in the six months after the lines were painted compared to the same six-month period the year before. Similar behavioural design nudges are now being applied in China and Israel to curb speeding.

In another trial in the UK conducted by Norfolk County Council, more than 200 trees were planted on the approach roads in north Norfolk which had a history of speeding problems. Results found that drivers reduced their speed by an average of 2 miles per hour. Again, as the car approached the village, the trees, planted closer and closer together, gave the impression that the vehicle was moving faster. This encouraged the motorists to slow down.

In another experiment in the US, the Virginia department of transportation painted zigzag white markings instead of the familiar straight dashed lines, to caution drivers approaching the road-crossing intersection used by pedestrians and bicyclists. They found that zigzag markings slowed average vehicle speeds and increased motorists’ awareness of pedestrians and cyclists. They also noted that the effects of the behavioural design didn’t wear off once motorists became used to it—they still slowed down a year after installation.

Building infrastructure like traffic signals doesn’t mean people will always follow them. But creating behavioural design nudges like displaying the seconds remaining for the traffic signal to turn green, is likely to reduce the number of people who break the signal. Such behavioural design takes into account that people are usually in a rush.

Rationally speaking, people shouldn’t be breaking signals because they wouldn’t be acting in their self-interest by putting themselves in harm’s way. But human behaviour is not rational. Drivers honk even when there is no way that honking could clear a traffic jam. Even when the signal is still red, there are drivers who honk. Therefore, rational ways of changing behaviour like educating people or creating awareness-based campaigns are ineffective. What’s effective at getting people to reduce honking is “bleep”—a red button on the dashboard of a car that beeps and flashes when the driver presses the horn. To switch off the red button, the driver has to press it. This behavioural design nudge breaks the habit of drivers’ honking because now each time drivers want to honk, “bleep” makes them deliberate whether they should honk or not. Bleep has been shown to reduce drivers honking by 61% in a six-month and 3,800km-long experiment in Mumbai.

Behavioural design needs to be applied at pedestrian crossings at traffic-signal junctions. At various traffic junctions, there are two signals in view—one signal placed just after the zebra crossing and the second signal on the other side of the junction once you’ve crossed it. That makes drivers keep inching forward, not stopping at the zebra crossing and thus not allowing pedestrians to cross. So to get cars to stop at the zebra crossing, only one traffic signal needs to be placed just before the zebra-crossing stripes begin, so that drivers have no option but to stop to get a view of the one and only traffic signal.

It’s time authorities stopped relying on ineffective money-draining campaigns, driver education and enforcement of laws. Instead, we should test simple, practical, scientific behavioural design nudges to improve road safety.

Behavioural Design & Road Safety (Mercedes Benz)

Posted on Published by Anand Damani

Behavioural Design & Road Safety

Our latest series of talks is how Behavioural Design can solve key road safety issues like accidents, speeding, honking, making pedestrian friendly traffic junctions, motorbike lanes and ensuring safety for all stakeholders. These talks are being done for Mercedes Benz. If we do get permissions we’ll try to upload the talks. Nevertheless we will write about the Behavioural Design nudges soon.

Designing effective public policy (Mint)

Posted on Published by Anand Damani

Designing effective public policy

This article first appeared in Mint’s editorial column ‘Their View’ on 21st Dec, 2016

Behavioural science should be used to design effective evidence-based public policy

For the most part, designing policy has meant passing a law, a sanction or penalty that imposes a fine or imprisonment to effect desired behavioural change or action. It assumes that the connection between law and actual behavioural is linear. It assumes that people are aware of the law, realize it applies to them, that people weigh the costs of breaking the law with the risk of being caught, overcome the temptations of the moment, in favour of willpower and self-discipline, and comply.

However, in spite of alcohol being prohibited in Gujarat, Nagaland, and Bihar, it is still readily available in these states, and has helped create a network of bootleggers, liquor mafia, spurious liquor, and a complicit police. There are fines for not adhering to traffic laws like honking unnecessarily or not stopping the car before a zebra crossing, but they are far from being effective in getting people to take the desired action. There is a fine for littering, but our roads are strewn with litter. Even when retailers charge for plastic bags, its consumption continues to grow. Fines and sanctions curb people’s fundamental right to choose and, therefore, are met with resistance and are often counterproductive.

A nudge, on the other hand, is a way of encouraging or guiding behaviour without mandating, instructing, sanctioning or monetarily incentivizing. It leaves people with the freedom of choice and yet guides them to act positively. Instead of shutting down choices, a nudge changes behaviour with a lighter touch, a more empirical and behaviourally-focused approach to policymaking.

A pioneer in designing effective public policy by using behavioural design nudges is the Behavioural Insights Team (BIT) or the “Nudge Unit” of the UK. It was started in 2010, headed by David Halpern and advised by Richard Thaler, with the backing of then Prime Minister David Cameron. Like any new idea, it had its sceptics in government. But over the first two years itself, it demonstrated the value behavioural science could bring in designing policy, based on empirical methods, that led to better outcomes, easier services for the public and most importantly saved government money.

BIT conducts dozens of experiments in the form of randomized controlled trials or rapid low-cost trials in areas such as healthcare, tax, energy conservation, crime reduction, employment and economic growth. Among some of its popular work is how it helped the tax department collect more taxes. BIT worked with the tax department to send out different versions of letters to people who owed tax to test systematically if changing the wording based on behavioural science literature would make a difference. They tested whether adding a single sentence such as “most people pay their tax on time” would boost repayment rates. And it did. By several percentage points, bringing in tens of millions of pounds. It’s because social norms are powerful in getting people to take action. Wording can make a big difference in behaviour change—imagine 3G, 4G and Wi-Fi being reworded as “radiation”.

Another experiment was about motoring fines. It showed that adding an image of the owner’s car to the ticket, captured by the camera, made the owner significantly more likely to pay unpaid car tax. In a third experiment, they encouraged people to insulate their lofts or attics. Insulation reduces heat loss, reduces energy bills and costs much less compared to the overall monetary benefits, yet people weren’t insulating their lofts. They tested two offers—an attic-clearance service and extra discounts. The attic-clearance scheme was more than three times more popular than extra discounts because the biggest issue was attic clearance rather than cost.

In the area of employment, getting the unemployed to think about what they could do in the next two weeks, instead of asking what they had done in the previous two weeks, significantly increased the number of unemployed who got work faster, trimming millions of days off benefits. In behavioural science, such a nudge is termed “implementation intention”.

In the area of pensions, employees now automatically joined the company-sponsored pension scheme by default but still had the option of opting out. So now the default was automatic enrolment rather than actively choosing to do so, making good behaviour easy. That led to more than five million new pensioners. Behavioural science studies by David Laibson, Shlomo Benartzi and other behavioural scientists show that changing the default beats financial education hands down.

Other BIT experiments have showed how simple behavioural design nudges can reduce carbon emissions, increase organ donations, increase quit-rates of smoking, reduce missed medical appointments, help students finish their courses, reduce discrimination and boost recruitment. And like the examples mentioned above, they are low cost, simple and scalable.

India has hundreds of problems to solve that require effective public-behaviour change—waste segregation, energy conservation, reducing road accidents, fuel conservation, cleanliness, adherence to medication, tobacco addiction, open defecation, reducing crime, hand-washing, tax evasion, alcoholism, etc. Instead of relying on law, fines, threats and monetary incentives, why not apply behavioural science and test simple, low-cost behavioural design nudges to see what works? Test, learn and adapt. After all, evidence-based policy is the best policy.

The science of how radio stations introduce new songs

Posted on Published by Anand Damani

The science of how radio stations introduce new songs

From Left to Right – Miley Cyrus, Beyonce, Grimes (Canadian Electronic Artist) Rihanna and Adele

As the major contributor of content on radio is music, the station ought to be playing hit numbers so that listeners keep tuned in. But how can radio stations afford to play only hit songs, if they are more costly? And what about the promotions of new songs? How do new songs get played on radio stations? And how do they become a hit? That got us digging. Here’s what we found.

There’s a company named Polyphonic HMI – a bunch of artificial intelligence experts and statisticians based in Spain – who has created a program called Hit Song Science. Don’t kill us for this. Hit Song Science analyses mathematical characteristics of a tune by comparing the tempo, pitch, melody, chord progression, and other factors against thousands of hit songs stored in Polyphonic HMI’s database. They predicted the success Norah Jones’s Come Away with me that won 8 Grammys, Santana’s Why don’t you and I that reached number three on Billboard charts. (We don’t know what song they predicted incorrectly.)

The program also predicted the success of a song called Hey ya by hip-hop group OutKast in 2003. When industry folks heard Hey ya, they liked it and thought it would become a huge hit. But when the data came in about listeners across radio stations, a huge number of listeners tuned out within the first thirty seconds itself. What went wrong?

One of the pioneers in studying why and how songs become sticky, Rich Meyers, says, “Sometimes radio stations do research by calling up listeners and playing a snippet of a song, and listeners will say, I’ve heard that a million times and I’m totally tired of it. But when it comes on radio, your sub-conscious says I know this song, I can sing along. Your brain secretly wants that song, because it’s so familiar to everything else you’ve already heard and liked.”

Areas of our brain that process music – auditory cortex, thalamus and superior parietal cortex – are designed to seek out patterns and look for familiarity. Music after all is complicated with numerous tones, pitches, overlapping melodies, etc. Our brains crave familiarity in music, because familiarity is how we manage to hear without becoming distracted by all the sound. That’s why songs that sound ‘familiar’ – even if we’ve never heard them before – get sticky.

The problem wasn’t that Hey ya was bad. The problem was that it wasn’t familiar. So radio stations used the trick of sandwiching Hey ya between familiar hit songs to mitigate risk. If stations don’t take risks with new songs people will stop listening. On the other hand, listeners want songs they already like. So by sandwiching them between already hit songs, stations make new songs familiar as fast as possible. Jiten, partner at Boing recording studio, says, “radio stations mitigate risk by playing new songs that are well promoted on audio-visual media like the TV as well.” Makes sense.

Moving towards Swachh Bharat (Mint)

Posted on Published by Anand Damani

Moving towards Swachh Bharat (Mint)

This article written by us appeared in the editorial section of Mint on 30th September 2016

A couple of weeks back, a video made by a private organization promoting the Swachh Bharat Abhiyan, featuring Kangana Ranaut and other Bollywood actors, went viral. The video depicted the picture of goddess Lakshmi disappearing from photo frames when people indulged in littering. The narrator on the video was Amitabh Bachchan, who said that the goddess of wealth lives only where there’s cleanliness. It ended with a plea by Bachchan and Ranaut to keep the country clean by not littering. Though the government didn’t issue this particular video, it has issued other, similar ad campaigns in public interest that promote the use of a public toilet instead of open defecation.

It is largely believed that ad campaigns change public behaviour by creating a change in people’s mindsets, which in turn leads people to take the desired action. But changing behaviour is not so easy. There are too many assumptions for this model of awareness leading to action.

The first assumption is that people can recall the message all the time. The second assumption is that the message is successful in motivating people to such an extent that it prompts them to act. The third assumption is that at the moment of actual behaviour, people would have the right amount of motivation, and also the ability to act in the desired way. That is a tough ask.

This is not the first time that the government has used ad campaigns to try and change public behaviour. In the recent past, campaigns like Save Fuel, Save Money have been aimed at changing driver behaviour by asking them to switch off car engines at traffic junctions to save fuel. Do you remember the campaign? If you don’t, crores of rupees in the form of advertising have been wasted. But let’s assume you are one of the few who do recall this message. Has it changed your behaviour? Do you now switch off your car’s engine at traffic signals?

Most people don’t. It’s a lot of effort. You need to turn the ignition off every time you wait at a traffic signal. And when the signal turns green, you have to turn the ignition on, listen to frantic honking because you haven’t moved immediately, change the gear from neutral to first if you are driving a manual-gear car, get frantically honked at again, put the hand-brake down, and finally get moving. Even for people who are highly cost-conscious or environmentally conscious, it’s too much effort.

That’s why campaigns are a money-draining and time-consuming way of attempting to change behaviour. In the UK, for example, in the 1970s and 1980s, the government spent millions on TV, radio and billboard ads educating people to wear seat belts. Researchers F.M. Streff and E.S. Geller estimated that by the end of the 1980s, 80-90% of British people had seen these ads eight-nine times each.

One would assume that showing people being launched head-first through their windscreens would make people respond. But it turned out that most people weren’t wearing seat belts. It was when the law changed in 1983, along with strict policing, that most people started wearing them.

Behavioural science suggests that a lot of the messaging on educating people to change behaviour seems powerful and emotional in the spur of the moment, but eventually doesn’t change behaviour because mere awareness rarely leads to action.

Changing behaviour is tough. People don’t always behave in the desired way. People should be exercising regularly, but many don’t. People shouldn’t be overeating, yet many do. The traditional way to change behaviour is to make people aware of the pros and cons of a particular act. But this method is ineffective, because most behaviour is instinctive i.e. subconscious. We aren’t always aware of the reasons for our actions. It takes a lot of time, effort and money to make someone aware of their behaviour, convince them that change is necessary and motivate them to change.

Behavioural science, on the other hand, uses subtle on-time nudges to enable the desired action. It focuses more on the ability to perform the desired action in the last mile than on motivating people. These nudges are based on a combination of behavioural economics, cognitive neuroscience and psychology. The nudges are designed to automate the desired action and for it to take place right at the moment of action.

For example, to reduce honking, we conducted an experiment in which a red button called Bleep was fitted on to the dashboard of the car. When the driver pressed the horn, the red button would begin to beep and flash. In order to switch it off, the driver needed to press the button.

The button made the driver conscious of the habit of honking by giving him immediate feedback in order to reduce indiscriminate honking. In a six-month experiment, Bleep reduced honking by 61% on average.

Similarly, a nudge was used in Copenhagen, with green footprints painted on the ground, pointing the way to the nearest garbage bin, that reduced littering by 46% by painting.

Meanwhile, to keep India clean, we first need dustbins that are easily accessible and cannot be stolen. They could be designed to include that extra bit of motivation for use—for instance, by having two sections and a question such as: “Who’s your favourite actress: Kangana or Deepika?”.

Clocky & Tocky – Alarm clocks that literally get you off your bed

Posted on Published by Anand Damani

Clocky & Tocky - Alarm clocks that literally get you off your bed

Have trouble waking up in the morning? Do you hit snooze on the alarm repeated like me, only to go back to bed and wake up later to realize that you are, as usual, late?

Here are two interesting products of Behavioural Design that could be of help. Clocky and Tocky are products created by Gauri Nanda, an MIT grad. Clocky is an alarm clock with wheels. When it’s time to wake up, these alarm clocks leap off your nightstand and run away beeping to ensure you’re awake. Tocky even lets you load your own wake up message or your favourite song.

See them in action by clicking here.

Related Posts Plugin for WordPress, Blogger...