Category Archives: Behavioural Economics

Self Control, Motivation and Why Bronze is better than Silver [Startuponomics 2011]

The second talk at Startuponomics was made by Dan Ariely and it concerned the problem of self control and how to go about motivating oneself better. A summary of the first talk is here – What Can Jam Teach you about Choice.

clocky alarm clock - Motivating?

He discussed people always wanting something in principle but typically not wanting to act to get that something right now. There are many things in life where we are designed not to care about the right things. Typically long term decisions, decisions far in the future, decisions that take alot of time, decisions that have high complexity and decisions that we don’t have much impact over in the short term.

He discussed his own personal problem of liver cirrhosis which required him to take a painful dose of medicine regularly for 18 months. He went about motivating himself to take the injections on time by pairing the taking of the medication with watching his favourite movies. He calls this principle reward substitution.

Would you prefer a half box chocolate now or a full box in a week? Most people say a half box now. Yet phrase it as a half box in a year or a half box in a 1 year and a week, and people switch their preferences.

Part of the success of the Toyato Prius over its competition was due to how it looked. It looked distinctly different. It gave people ego utility. People were able to signal to others that they cared about the environment whereas this was not possible with other cars as they looked the same. This illustrates that making some things visible can improve behaviour. It shows that finding other types of motivations can make a difference.

Find other types of motivations. We care about competition, completion, image, accomplishment. These things motivate. Money, Guilt, Shame, Social proof, Ego.

Thinking about the global warming problem. It is a very hard problem to solve. This is because it is far in the future, it is uncertain, it will likely not ultimately affect us,   it requries a change in entrenched behaviours and individually we can have very little impact on it.

Study – How do you go about getting patients that have had a 2nd stroke to take their medication? Apparently informing them that they have a high chance of death if they don’t take it is not enough. Some ideas are to notify kids if you don’t take it or give a reward if you take the medication. Sometimes a gift is more motivating than cash of the same value. You could also make something disappear if they stop using it, eg mobile phone not working. You could use fear of dying. you could create social connection to other people taking the medication. You could show a picture of someone getting fed today in a poor country if you take your pill (and not give the idea that they child will not be fed if you don’t). What worked best was letting people participate in a lottery related to the taking of their pill. On the first day they didnt take it, they were informed that they had won the lottery. They were then informed they wouldn’t be awarded the prize as they had not taken their pill. Taking the medication increased significantly.

Think about how long we need to sustain the motivation and what happens if we stop.

There is a point where giving money makes people moitvated (e.g. giving $1000 to floss your teeth, flossing would increase for most people).

A 1/1000 chance to win a thousand dollars is felt as more motivating than the same $1 value of the chance.

Regret is the emotion where you imagine another state of your life.

Study – Gold, bronze, silver example at the olympics. Gold and bronze are observed as the happiest. Why? Silver imagines themselves in the gold state and bronze is just happy to have avoided 4th.

Self control contracts are powerful. Ulysses and the sirens example. He knew he would be tempted so he tied himself to the ship.

Study – Rats were trained to press two buttons. One was the press a green lever and get a pellet of food. Another lever was to wait ten seconds and get ten pellets of food. Rats often chose the instant gratification of the first lever. But when a no temptation lever was added which disabled the green lever, some rats were able to press it. Implication is that imposing self control contracts can be powerful.

Examples of innovated product design using these principles – We are a different person at night compared with 6am in the morning regarding aiming to go for a morning run. the product called Clocky (hard to find alarm, have to find it to turn it off). Alarm clock connected to bank account and charity you hate. Imagine a website that is connected to your parents that tells them if you are watching porn.

People need red buttons. Need to encourage self control.

Not always about habit, it is about what decisions we can make that are sustainable.

Dieting vs no smoking example. Need different motivations for different goals.

Contract example of sending letter to mum about heroin use. You agree to send your parents a letter, write the letter and give it to a company that is bound to send it if you fail a drug test at any time in the future (tests are random). This idea was stopped as it created violations of human rights but it was also effective.

What Can Jam Teach You About Choice? [Startuponomics 2011]

From the people who brought you the Intelligent Investor’s 15 day free trial series, today the Startuponomics trip report series has begun. Be excited, be ready.

I recently attended Startuponomics in San Francisco, a conference run by Dan Ariely on behavioural economics, decision making, irrational thought, motivation, influencing behaviour, and how people think. I learned a huge amount of interesting information at this conference and I hope this little series will help share some of it for those interested. Some posts will be longer than others as they are summarising 1-3 hour discussions. Subscribe to my rss on the right to ensure you don’t miss out! Click here to unsubscribe.

Now, let’s get on with it.

Choice

Dan Ariely – Day 1

Think of behaviour as a flow or process. Where can we intervene?

He discussed external vs subjective reality. Your intuition about how the world operates is biased.

Study – Table length illusion example. E.G. told visual illusions are X, but our brain keeps seeing Y.

Example about box checking. Check the box if you don’t want to participate. Opt in vs opt out. The default is to not check the box no matter what the question is. Consider the fact that we are less active than we think in making our decisions.

The environment of the interface has a massive impact on the customers decision. If questioned, we come up with stories about what we chose even though our reasons may not be real.

Study – Long term medication company. Drug rep example. generic vs branded medication. Trying to make people do something different to what they’re doing. 80/20 branded / generic went to 10/90 after telling people that they had to choose one or the other compared with asking them to choose generic which caused little change.

When discussing it, people lie or get it wrong. People behave as if it is expensive to return the postcard when it would save them money. Understand how defaults work. Question about assuming change and if people still didn’t respond.

Study – Physicians example where physicans were told that they had made a mistake a day prior with a patient – would they call them to inform them to take a pain killer? 60/40 in group 1. All did nothing in second group when decision was nothing, ibuprofen or some other drug. Defaults are important. But when you add more complexity to the decision to do nothing or not, more people do nothing. [Ask Question about competing principle economist example.]

Study – He gave an example about reducing the urinal bed pans in hospital (maybe do it by changing the default). Default matters alot in things that we pay alot of attention to. As things become more difficult we want to do less and as a result interface designers decide for us.

Study -The jam study. 6 or 24 Jams. 40/60 approach the table in each case respectively. 1.4/1.5 try jams. 30% /3% actually buy. ” I’ll do it next time” is more common amongst the high choice option. Excitement can be overturned by too much choice. Technically more options fits more tastes but also results in confusion. Options increase utility up to a point then it goes negative. Then it reverts to the default of nothing if too much choice.

Study – 401 k example. A ways to make less contributions is to make it opt in rather than opt out. Give lots of options. Make the decision seem important. [Ask question about what effect importance has]. Think about how much time we actually spend on important decisions. Not enough (marriage kids university).

Switch with opt out option works.

Lots of software is resistant to changes. Think about how we build software as a result of irrationality.

Defaults discussion 

Default is no mail delivery. If we want more people on mail, we should change the interface.

Default is click here to not receive offers from us.  Similarly, this can be manipulated.

Default is not having a free trial or access to an article. One idea is to sign them up automatically and tell them to enter their details to complete the process.

Can you think of any more?

Zero has a special place in people’s hearts. Don’t train people to think the service costs nothing. Give it a value. Make people work for it, fill out a survey to get it free might help people value their trials at ii (though it also may reduce sign ups). We can do this after we have their email address.

Study – Cheap airline example where a long one step interface with matrix style options was made into a two step process with other changes. The idea to break it into two screens and changed the previous premium option to basic and the others to basic (minus x,y, z).  Very successful (so much so that results not revealed). Used loss aversion in the process. How to trial this on our site?.

We have lots of individual differences but are tuned to the differences. We are all more similar than we think.

There is a trade off between how many people finish versus the quality of the data.

Study – The pre holed card for caraway example. Pre-punched cards (10 car washes to get one free, 20 to get one free with 10 pre-punched) – people think they’re half-way through and complete the second lot more.

Study – The WordPress.com example… Should they create a new step in the sign up process to let people choose their theme. Work hard before you do anything else. Ikea effect.

If you believe in people being irrational, how do you feel about forcing people to do things?

Influencing decisions once versus habits. The 3 vs 10 reasons you love someone example. Participants asked to list either 3 or 10 reasons that they love someone. The way we are made to think about something for time can change our opinion since this changed how much they said they love the person when surveyed afterwards. The extent to which we don’t know our preferences is quite large. Interface makes the difference. In this case 3 was more powerful as people couldn’t think of 10 reasons often.

E.g. Think about 15 ways this class could be improved. What if you do it in week 4 vs 12, they forget. time lapsed is crucial.

Study - How often do you floss example. Giving a scale can influence their future depending on where on the scale they fall (so you can manipulate where people will fall on a scale and how they will act by varying the scale). Then more people went to dentist when low on the scale. Scales can Shape behaviour.

Study – Financial advisor risk attitudes example. Asking people to remember bad things before asking preferences can have an impact. “stock market example”. Can manipulate emotion in this way. But once answered they take it very seriously. They will then rely on it.

When market research about what they think they think through surveying they will make it up on the spot. The way you ask the question really influences the answer. Observing preferences is very different to directly asking for them.

Computer processor example. Audi configuring example. People don’t actually know they preferences for buying cars and the average car bought was always close to the default options (several different default car options were tested (e.g. base including leather seats / base not including leather seats).   People don’t know their preferences. Need to get people to committ while they are primed.

I’m not certain I’ll continue the series on this blog (i was only planning on doing via email), so if you are interested in hearing more about the talks, please like this post or post a comment below.

Edit – The next article in the series is available here – Self Control, Motivation and why Bronze is better than Silver