Most darts players aim for the treble 20, because that’s what the professionals do. However, for all but the best darts players this is a mistake: if you are not very good,* your best approach is not to aim at treble 20 at all, but instead to aim at the south-west quadrant of the board, towards treble 19 and 16. You won’t get 180 that way, but nor will you score 3. It is a common mistake in darts to assume that you should simply aim for the highest possible score – you should also consider the consequences if you miss.
Many real-life decisions have a scoring rubric that is more like darts than archery. For instance, in deciding whom to marry, aiming for the best may be less important than avoiding the worst – rather than trying to maximise an outcome, you may seek a pretty good all-round solution with a low chance of disaster. A darts player repeatedly aiming for the south-western quadrant of the board would look insane to many onlookers, who might say, ‘You’re supposed to aim for the triple 20 – it’s the highest score on the board,’ but an approach seeking to minimise variance or minimise downsides often involves behaviour that seems nonsensical to those who don’t understand what the actor is trying to do.
By the same token, to someone who assumes that holidaymaking is a lifelong quest to find new experiences, returning annually to the same resort may seem ridiculous; it is on the other hand an extremely good approach if you want to avoid a bad holiday. Habit, which can often appear irrational, is perfectly sensible if your purpose is to avoid unpleasant surprises.
Social copying – buying products or adopting behaviours and fashions that are popular with others – is another safe behavioural approach. After all, the bestselling car in Britain is unlikely to be terrible. Another reliable risk reduction strategy when making decisions under uncertainty is simply to substitute a different question from the one that conventional logic assumes you should be asking. So you would not ask ‘What car should I buy?’, but ‘Whom can I trust to sell me a car?’ Not ‘What’s the best television?’, but ‘Who has most to lose from selling a bad television?’ Or not ‘What should I wear to look great?’, but ‘What’s everyone else going to be wearing?’
A common approach when recruiting staff is to ask your existing employees to recommend people – indeed, this is how most entry-level jobs in mid-sized companies seem to be filled. This might seem to involve fishing in a very narrow pool, and it is, but a personal recommendation from an existing staff member is a good way to avoid hiring someone terrible. People are keen to do favours for their mates, of course, but no one is going to jeopardise their reputation at work by recommending an alcoholic, a kleptomaniac or an arsonist. Third-party recommendations are not perfect or remotely scientific, but they are rarely catastrophic.
Many apparent paradoxes of consumer behaviour are best explained by similar mental mechanisms. A few years ago we discovered that men were reluctant to order a cocktail in a bar – in part because they had no foreknowledge of the glass in which it would be served. If they thought there was even a slight chance that it would arrive in a hollowed out pineapple, they would order a beer instead. One remedy was to put illustrations or pictures of the drinks on the menu; some trendy venues have since solved the problem by serving all their cocktails in mason jars. The same sort of mental calculus explains why it is so difficult to get people to move their current account from one bank to another paying a higher rate of interest, or to shift their broadband provision. A 1 per cent chance of a nightmarish experience dwarfs a 99 per cent chance of a 5 per cent gain.
5.6: JFK vs EWR: Why the Best Is Not Always the Least Worst
I once asked, over Twitter, whether there were any clear advantages to flying to JFK Airport in New York rather than Newark.* Other than a string of replies from New Yorkers with an inbuilt disdain for anything in New Jersey* there seemed to be few arguments for using JFK: Newark is closer to Manhattan, and risks fewer roadworks or delays on the journey. Richard Thaler, one of the world’s most eminent decision scientists, tweeted me with strong support for Newark.* If it were only informed consumers making the choice, Newark would surely be the more successful airport, yet JFK is the more common choice. Ironically it may be more popular simply because it is more popular – if that sounds like nonsense, bear with me.
Because JFK is more popular, it is seen as a less eccentric choice. Flying to JFK is the equivalent of buying an IBM mainframe in 1978: an easy default. The great thing about making the ‘default’ choice is that it feels like not making a decision at all, which is what businesspeople and public sector employees tend to really like doing – because every time you don’t visibly make a decision, you’ve ducked a bullet. Newark requires a rational justification: ‘Why is my flight going to Newark; why not JFK?’ By contrast, the sentence, ‘I’ve booked you a flight to JFK’, rarely meets with the question, ‘Why JFK? What’s wrong with Newark?’
So, imagine you are the personal assistant to a grumpy boss in London. He or she asks you to book a flight to New York. You have two choices.
Book the boss on a flight to JFK, hand over the tickets and relax.
Book them on a flight to Newark and cross your fingers.
There’s a large chance that if you take option 2 – the better decision – you will come out quite well. If your grumpy boss notices the ease of the journey from Newark and the friendliness of immigration staff, they may remember to thank you on their return; you might even get ‘Nice choice – remind me to use Newark next time.’ However, it is unlikely that your boss would buy you a case of vintage champagne and immediately award you a four-figure bonus – a thank-you is the best you can get.
But flights are delayed or cancelled – and the reason that you would need to keep your fingers crossed after choosing option 2 is that, when things do go wrong, as they sometimes will, the difference between choosing options 1 and 2 becomes more stark. If the flight from JFK is delayed by three hours, your boss will blame the airline, but if the flight from Newark is, your boss will probably blame you, because with option 2 you made a noticeable decision – you deviated from the default. He might say, ‘This wouldn’t have been a problem if you had booked me from JFK – the flights were fine there. What were you thinking, booking me a flight from this weird airport, you fool?’
Blame, unlike credit, always finds a home, and no one ever got fired for booking JFK. By going with the default, you are making a worse decision overall, but also insuring yourself against a catastrophically bad personal outcome. In his book Risk Savvy (2014), the German psychologist Gerd Gigerenzer refers to this mental process as ‘Defensive Decision-Making’ – making a decision which is unconsciously designed not to maximise welfare overall but to minimise the damage to the decision maker in the event of a negative outcome. Much human behaviour that is derided as ‘irrational’ is actually evidence of a clever satisficing instinct – repeating a past behaviour or copying what most other people do may not be optimal, but is unlikely to be disastrous. We are all descended from people who managed to reproduce before making a fatal mistake, so it is hardly surprising that our brains are wired this way.
In institutional settings, we need to be alert to the wide divergence between what is good for the company and what is good for the individual. Ironically, the kind of incentives we put in place to encourage people to perform may lead to them to be unwilling to take any risks that have a potential personal downside – even when this would be the best approach for the company overall. For example, preferring a definite 5 per cent gain in sales to a 50 per cent chance of a 20 per cent gain. Why else do you think Management Consultancies are so rich?
Part 6: Psychophysics
6.1: Is Objectivity Overrated?
You may have never heard the term ‘psychophysics’, which is essentially the study of how the neurobiology of perception varies among different species, and how what we see, hear, taste and feel differs from ‘objective’ reality. For instance, different species, as I will soon explain, perceive colour very differently, since receptors in the eyes ar
e sensitised to different parts of the light spectrum. More importantly, our different senses – though we don’t realise this – act in concert; what we see affects what we hear, and what we feel affects what we taste.*
A few years ago, the British chocolate manufacturer Cadbury’s received a large number of customer complaints, claiming that they had changed the taste of their Dairy Milk brand. They were at first baffled, because the formulation hadn’t been altered for years. However, what they had done was change the shapes of the blocks you would break off a bar, rounding their corners. And smoother shapes taste sweeter. Truly.
Nothing about perception is completely objective, even though we act as though it is. When we complain that a room is hot, there may be no point at which we agree about what ‘hot’ means; it may merely mean ‘a few degrees warmer than the room I was in previously, to which I have become acclimatised’. ‘Time flies when you are having fun’ is an early piece of psychophysical insight. To your watch, an hour always means exactly the same thing, regardless of whether you are drinking champagne or being waterboarded. However, to the human brain, the perception of time is more elastic.*
In some businesses, psychophysics is a more valuable discipline than physics, and in many industries you need to master both. The airline industry is a good example: along with the physics of flight, you need to understand the psychophysics of taste, because food tastes very different at altitude, meaning that meals that are pleasant on the ground can be boring in the air.* Endless complaints about airline food, which were once a staple of stand-up comedy, may have been unfair: it wasn’t that the food was bad, but it was the wrong type of food to eat at 30,000 feet.
The new Boeing 787 Dreamliner is, in many ways, a triumph of psychophysics. Lighting, pressurisation and humidity all mitigate the effects of jet lag. Moreover, visual illusions – in particular a spacious entranceway – create an impression of spaciousness; it is actually 16 inches narrower than a Boeing 777, but to many passengers it feels significantly wider. Adding a little space when people enter the aircraft creates an impression of airiness that carries through into the main cabin, even though the main cabin is no less densely packed than usual. Blake Emery, the psychologist in charge of product differentiation at Boeing, explains that his team were ‘looking for things that people really couldn’t articulate’ that might improve the experience. No one actually knows the humidity and air pressure inside an airliner, but these things have a large effect on how people feel. Historically, airliners were designed with the airline’s accountant rather than the passenger in mind – all they cared about was cost and capacity. So Boeing’s attempts to differentiate the passenger experience were a bold move.*
Engineers and accountants are prone to ignore the human side of their creations, and this is not always wrong – if you are building an unmanned space rocket or a bridge, it is possible to define success in objectively defined measures. However, if you are designing anything where human perception plays a part, you need to play by a different set of rules. For instance, a bridge that supports a prespecified volume of traffic and survives in all reasonable climatic conditions can be said to be a good bridge. It would, of course, be better if it were an attractive bridge rather than an ugly one, but that is a secondary consideration – in designing a bridge, there is little scope for alchemy.
In other projects, like designing a train service or a tax system, or in painting the lines on roundabouts, it is impossible to define success except in terms of human behaviour. Here there is generally some potential for alchemy, since perception, rather then reality, is what determines success. Even giving a tax a different name can have a colossal effect on whether people are willing to pay it.*
6.2: How to Buy a Television for Your Pet Monkey
You probably aren’t aware of this, but your television is cheating you. The same applies to the computer screen* and also to the colour pictures in magazines. Not everything on an LCD screen involves deceit: when the screen shows pure blue, green or red, it is more or less telling the truth – blue lights produce pure blue photons, green lights produce green photons and red lights produce red photons. Each pixel on the screen contains three LCD lights – one in each of those three colours. If the red lights alone are displayed, the screen is red.
But yellow on television is a big fat lie. It may look yellow, but it isn’t really – it’s a mixture of red and green light, which hacks our optical apparatus to make us think we are looking at something genuinely yellow. The yellow is created in our brains, not on the screen. Colour mixing is a biological, not a physical phenomenon – you can’t mix green and red photons to make yellow photons, but by sending an image of a mixture of red and green photons to the brain in the right ratio, the resulting stimulus is indistinguishable from that of yellow photons, and you see yellow as a result. Even so, yellow isn’t as much of a lie as purple* is – yellow does, at least, exist on the light spectrum.* Purple does not exist at all: indigo and violet are in a rainbow, but magenta isn’t – the colour exists only in our heads.
The reason for all this is that humans – and indeed all apes – have trichromatic vision. We have three sets of cones (or colour sensors) in our retinas, each of which is sensitive to a different part of the colour spectrum; the brain then constructs the rest of the spectrum by extrapolating from the relative strength of these three. In the case of purple, which occurs when the red and blue sensors but not the green ones are triggered, the brain creates a colour to fill the gap.* Three colours are hence enough to recreate a wide (though not quite complete) spectrum of colours – not on the screen, but inside our heads; including some colours that aren’t really there at all.
Because colour mixing is a biological phenomenon, how it works depends on the species (and sometimes the individual) which performs it. If lemurs and lorises bought televisions, it would be advisable to produce a cheaper dichromatic LCD TV for them – these primates construct their colour spectrum from green and blue alone, so you could omit the component of each pixel that generates red light.
Copyright © worldlifeexpectancy.com
Many female marmosets see in three colours, but males can only see in two.
It’s a good thing that marmosets don’t buy televisions, as it could be a source of marital discord. Females and males of the species have completely different colour perception – many females see in three colours, while males only see in two. A female might come home having spent £800 upgrading their 65-inch OLED to one of the new hyper-realistic tri-chromatic models, but her mate would complain that ‘it looks just the same as the old one’.* Your best bet is to keep a pet owl monkey, because they would be perfectly happy with a black-and-white TV: like many other nocturnal mammals, they don’t see colours at all.
No one advertises televisions as being ‘designed for higher primates’, but it would be perfectly accurate to do so. The lesson to take from this is that it is possible for something to be objectively wrong but subjectively right. TVs are designed around how we see, not what they show. There is a lot of clever engineering involved in making a television,* but the real genius in it is psychological alchemy, not technology – without an understanding of how humans perceive colour, making one would be almost impossible.
As I have argued, psycho-logic and psychophysics need to be applied not just to the design of televisions, but also to welfare programmes, tax, transportation, healthcare, market research, the pricing of products and the design of democracy. There is no point in struggling to create changes in objective reality if human perception can’t see it, so all these things need to be perception optimised for humans. Moreover, just as with the colour purple, we should remember that if you design something in a certain way, people can perceive something which doesn’t exist in reality.*
What really is and what we perceive can be very different.
This is where physical laws diverges from psychological ones. And it is this very divergence which makes Alchemy possible.
6.3:
Lost and Gained in Translation: Reality and Perception as Two Different Languages
There is a whole academic discipline devoted to the idea that human behaviour can be modelled as if it were a physical phenomenon: it’s called economics. However, the simple fact is that, in all facets of human behaviour, reality and perception are like two different languages, each with concepts that are more or less untranslatable into the other.*
Emotions are stranger still and, like magenta, are produced in the mind. If a tree falls in the forest and no one hears it, does it make a sound? Yes, because mechanical sensors could still record that sound. But if a car lingers too long at a green traffic light and there is no one waiting behind to get angry, is it an annoyance? No, because annoyance is a perceptual concept that is confined to living things. Obviously, perception and reality sometimes closely parallel each other, but at other times there is a complete disconnect between the two that is similar to a language gap. If you take any two languages at random, you may find that the two are sometimes immensely different, containing concepts unique to one or the other.* Other language-pairs are similar,* perhaps confusingly so, but each situation brings its own problems, and in both cases you can experience translation errors.
Alchemy Page 21