Designing an Optimum Nudge

I’m sitting by a window looking out at a rainy Paris street, thinking of cultural differences between Paris and San Francisco, taking advantage of bad weather to do some writing. Over two decades ago, I did some ethnographic research a Exploratorium, looking at how different visitors interacted with the museum’s hands-on exhibits. I was looking for ways to improve the visitors’ experience, raise understanding of the phenomena they were observing. What I saw was different ways in which visitors experienced failure: p-prims that got in a way; folksy wisdom that caused confusion; lack of affordances that led to bottlenecks; permission giving that set up strange expectations; etc. The results of this study turned into a Master Thesis for UC Berkeley.

Now, I would like to explore some of the ideas that surfaced during my Exploratorium research and apply them to design of nudging — carefully crafted affordances and perceptual cues that manipulated users into acting a certain way while maintaining the illusion of freedom of action. Let me start with a bit of history — a quick summary of some of the results of Exploratorium study.

Permission Giving

Tornado Exhibit at the ExploratoriumTwo decades ago, “hands-on” exhibits were still novel in the museum world. Exploratorium in San Francisco was the pioneer in this approach, but the cry for active participation wasn’t taken up right away. In particular, visitors coming from Europe, where museums were places of stored artifacts too fragile and valuable to touch, were very hesitant to haptically engage with the museum’s offerings.

I repeatedly observed parents pull their kids away from coming into contact with Exploratorium’s “artifacts”. What if they broke it? Where are the guards that yell at kids to stay away? Exploratorium was a whole new way of exploring the world, and visitors needed to be told explicitly what was okay and what wasn’t — to be given permission to act in ways that were counter to their previous experience and judgement.

There are many ways to give permission to act a certan way:

  • Signals — red and green traffic lights are great examples of signals that allow or forbid a certain action.
  • Signs — “don’t touch” or “don’t feed the animals” are common ways of letting people know what is forbidden; but there are few signs that specifically allow something: “please touch”! Most signs are about limiting permission or limiting actions (even ones that tell you to do something: “turn on the light” and “sit down”). Signs can be linguistic or visual or audio-based.
  • Affordances — physical design constraints that encourage a particular action. While the following examples are extremely simplistic, they get the gist: don’t place chairs in a waiting room if you want people to stand; the light switch next to light fixture is probably the one; a knife has a business end and a handle and it is easy to guess which is which…
  • People in Authority — people we recognize as authority giving us explicit permission to act or do something. A uniform is usually a good sign that a person has powers over us.
  • Example — we can watch others do something and take their actions as permission to do the same. This is one of the more interesting ways of getting permission. We are group animals. We try to conform to group norms. We hate to stand out, especially in unfamiliar situations. To figure out what’s okay and what’s forbidden in a novel place, we stand back and watch what others do — learn from “old timers”. We can learn appropriate behavior in real time or by doing research into cultural norms of a place we are about to visit.

Above is a photo of a Tornado Exhibit, one of the more popular exhibits at the Exploratorium and one which is now included in many other museums around the world. I have observed countless instances where visitor to the museum stood back and watched some adventurous kid jump inside and start walking around the tornado. I’ve even modeled the behavior myself when the museum was too empty and didn’t provide too many permission giving opportunities.

But museums are not the only places for permission giving. Consider this TED video featuring Paul Bennett:

The talk gives a multitude of permission giving examples — nudges that motivate individuals and groups to act in certain, desired ways.

Herd Mentality

Yesterday, we’ve spent a nice day reading at the Jardin de Luxembourg. There was a respite from rain and people flocked to enjoy the day in the beautiful garden. Unlike the parks in U.S., the lawns at the Luxembourg garden were perfectly manicured, clean, and untrodden. No one but the the tiniest toddlers (minus the shoes) ventured onto the grass. And because no one was doing so, there was no permission for anyone to do so. In fact, it felt like if one put a foot onto the grass, there would be stares of rebuke from all around. Now, there are other lawns in Paris that are used to lounge on, just not those. And without a single sign or notice of “no trespassing”, there was not a single person breaking the rule. And this was true for people who were in the garden for the first time — lots of tourists wondered in and out all day. And without prior knowledge of notification, people obeyed.

The force of the herd is strong!

Instilling Good Behaviors

I’ve lived in New York for over a decade. Water shortages were not an issue, and everyone I knew there ran the water while brushing their teeth. It was not until I’ve moved to California, a desert country, that I started to pay attention to water. The civic messages to conserve this precious resource were everywhere! Still, it took years before water conservation became an ingrained habit. What’s that? It’s when you are made to feel very uncomfortable at the very thought of running water! Water conservation stopped being a conscious decision and became an subconscious emotional preference.

In Paris, water is not a problem. I routinely watch as streets are flooded on purpose — for cleaning. And as I watch the water flow, I cringe. It makes me uneasy to watch so much water wasted so freely. It makes me want to turn it off. I find it difficult to walk by… (here’s a link a previous post where I talk about water use in Rome)

But for how long? How long will this discomfort last? Watching the water flow, and observing total acceptance from all around me at this waste dulls the feelings of discomfort. At some point, if I continued to stay and live in Paris, I would stop noticing the waste, too. And, I believe, it would take far less time for the discipline of conservation to cease than for it to develop in the first place.

Habits take time to form. Good habits — those requiring disciple or restraint — are harder to instill and maintain than bad ones. It’s much easier to drop the dirty socks on the floor than to put them in the hamper; it takes less energy to stockpile dirty dishes in the sink than to wash them. Some actions require mental discipline because they require energy and thought in their execution. I could just dump the used napkin on the street, but I look for a waste receptacle. I feel uncomfortable, even guilty, throwing things on the ground — a good habit I’ve developed. But for how long? What would it take for me to litter? This is where design comes in!

A Few Observations on Nudge Design

Proximity and Convinience — I’m much more likely to properly dispose my used napkin if there’s trash-can near by. If we want people to do “the right thing”, we have to make it within reach and visible. If I don’t see a proper way of discarding waste, then I have to find ways of managing the waste on my person until I find the right receptacle. This works for “clean” and “small” waste, but it progressively gets worse. Consider dog poo. How long will the owner of the dog be willing to carry the poo before just dropping the container over some fence (or some other random place)? Municipalities around the country (and in France) try to save money by removing trash bins, the result is more trash that is harder to clean! My neighbor removed over 50 dog poo bags that people thrown into her hedges (she just moved in and wanted to do some plantings, when she discovered the scale of poo garbage in her new yard). These were not all bad people: they’ve collected the dog poo as they walked their dogs, but there was not a convenient disposal location — so over the fence it went!

Similar problems result in lack of public toilets, and lack of toilets that are safe and clean to use. Last night, as we were walking through Jardin de Plants, an elderly gentlemen in a suit and tie stepped over a little rail guard and walked to the tree and whipped it out and urinated in public! There were people everywhere, but no one even looked (unspoken permission by the herd) or said a word or raised an eyebrow (well, I did!). The park is full of kids and this kind of behavior in San Francisco would get a mom at the playground calling the police… But not in Paris. There are toilets in the Park, but they are distant (proximity problem); you have to pay to use (not convenient); you have to wait in line (the capacity is not large enough); and by watching a woman roll up her pants prior to going in, I deduce that they are dirty. All these factors nudge Parisians into “bad behavior”. [Paris by no means is the only city where this happens. I grew up in Leningrad, and public toilets were unheard of… with predictable consequences.]

Group Norms — The above public urination example is a classic example of group norms. I discuss the issue of Parisian men and public urination as a design problem: how can this be fixed by design? Certainly the society as a whole would improve if men stopped urinating on every corner (yes, this is a very big problem in Paris). The smell, the sanitation issues, all bring the public discipline down.

And consider graffiti: the first instance probably required a lot of nerve from the offender, but later, it’s not a big deal — everyone is doing it — you get a permission by observation, by example. That’s why neighborhoods try to eradicate tagging as soon as possible, prior to instilling undesired habitual behavior. In Russia in the 60’s and 70’s, graffiti resulted in a prison term with a high probability of being beaten near death by the police.

Punishments and Rewards — In San Francisco, it is the responsibility of the owner to keep his property free of graffiti. For every day the marks are there, the fines go up! Even though the owner is a victim of the crime, graffiti is judged so detrimental to the neighborhood as a whole, that it is acceptable to penalize the victim to prevent escalation. (I once spent a whole day trying to remove a tag from the pavement in front of our house — nasty chemicals…)

Punishment is of course not the only way to engrain a desired behavior. Rewards work just as well, even better! Again in San Francisco and other Bay Area cities, home owners are rewarded with money for installing solar panels (although it is still too expensive to put up; doesn’t do much good in the fog; and requires motivation beyond the government payout…).

Ease of Execution — Consider the need to conserve electricity. A noble goal. But what if a person leaving the building last had the responsibility to make sure that all lights were out? If that person had to run around the whole building to ensure the task is done, the lights would be habitually left on. Simply other responsibilities are more important: the need to pick up the kids from school of time, for example. How much time is it worth to deal with the lights? But if there was a central switch — one click and it’s done — then more people would conserve electricity. It’s easy to do and people feel virtuous doing so.

External Monitoring — There’s a great story of a hospital trying to enforce its doctors and nurses to wash hands after leaving the bathroom — yes, it’s a problem! The solution was to post a sign: “Make sure that the people around you washed their hands.” Just asking to wash hands didn’t do the trick. But making people feel watched, monitored, improved compliance — more clean hands, better survival rate for patients.

Self-benefit — In theory, self-benefit is a huge motivator. But how many people take all their medication on time? If it is not convenient, easy to access, simple to use, or financially feasible, people go without. The benefit of longer life and healthier life is just too far in the future. Rewards and punishments have to be in direct proximity in time and space. This is behavior therapy one-oh-one.

Temptation and Fun — A push to do good can be achieved with entertainment. Consider this video:

Humans are curious animals. We want to look around the corner. We have to teach ourselves not to peak at the back of the book to find out how things turn out. We will open draws and medicine cabinets in our friends’ houses… even when we know it is wrong… but perhaps we won’t get caught? [I never do this myself, of course.] Thus temptation, curiosity, and fun can be used to nudge people towards desired behavior.

Designing an Optimal Nudge

Hard things — like throwing away garbage, not peeing in public, picking up and disposing dog poo, not spitting on the street (read about China’s efforts to eradicate public spitting prior to hosting Olympic Games) — take discipline. Discipline takes a lot of support: the right motivators, punishment and rewards, monitoring, well-designed affordances, and so on. New group norms take time. The harder the change, the more cultural support the old behavior has, the longer it will take to engrain the new behavior in a population. And it is much, much easier to slip back into doing what’s easy (as opposed to what’s right and hard). Get rid of trash cans, and even the most motivated dog owners will talk themselves into throwing the “little presents” into their neighbors’ yards rather than taking them into their cars and homes.

I have more to say on this subject… but that will have to wait for the next post…